diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_and_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_and_.md new file mode 100644 index 00000000000..a6a9647868d --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_and_.md @@ -0,0 +1,18 @@ +## [ 仅参数名不一致 ]torch.Tensor.bitwise_and_ +### [torch.Tensor.bitwise_and_](https://docs.pytorch.org/docs/stable/generated/torch.bitwise_and.html) +```python +torch.Tensor.bitwise_and_(other) +``` + +### [paddle.Tensor.bitwise_and_](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor__upper_cn.html#bitwise-and-y-out-none-name-none) +```python +paddle.Tensor.bitwise_and_(y, name=None) +``` + +两者功能一致，输入参数用法不一致，具体如下： + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| other | y | 按位与的另一个输入。| diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_xor_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_xor_.md new file mode 100644 index 00000000000..308a91aaec4 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.bitwise_xor_.md @@ -0,0 +1,18 @@ +## [ 仅参数名不一致 ]torch.Tensor.bitwise_xor_ +### [torch.Tensor.bitwise_xor_](https://docs.pytorch.org/docs/stable/generated/torch.bitwise_xor.html) +```python +torch.Tensor.bitwise_xor_(other) +``` + +### [paddle.Tensor.bitwise_xor_](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor__upper_cn.html#bitwise-xor-y-out-none-name-none) +```python +paddle.Tensor.bitwise_xor_(y, name=None) +``` + +两者功能一致，输入参数用法不一致，具体如下： + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| other | y | 按位 y 异或的另一个输入。| diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.cumsum_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.cumsum_.md index d016b7e5a3e..fe20428e59a 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.cumsum_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.cumsum_.md @@ -1,7 +1,7 @@ ## [ 仅参数名不一致 ]torch.Tensor.cumsum_ ### [torch.Tensor.cumsum_](https://pytorch.org/docs/stable/generated/torch.Tensor.cumsum_.html) ```python -torch.Tensor.cumsum_(dim, dtype=None) +torch.Tensor.cumsum_(dim, *, dtype=None) ``` ### [paddle.Tensor.cumsum_]() diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_and_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_and_.md deleted file mode 100644 index 0761866182b..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_and_.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.Tensor.logical_and_ -### [torch.Tensor.logical_and_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_and_.html) -```python -torch.Tensor.logical_and_(other) -``` - -### [paddle.Tensor.logical_and_]() -```python -paddle.Tensor.logical_and_(y) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| other | y | 表示输入的 Tensor ，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_or_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_or_.md deleted file mode 100644 index 6f29d2519c4..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_or_.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.Tensor.logical_or_ -### [torch.Tensor.logical_or_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_or_.html) -```python -torch.Tensor.logical_or_(other) -``` - -### [paddle.Tensor.logical_or_]() -```python -paddle.Tensor.logical_or_(y) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| other | y | 表示输入的 Tensor ，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_xor_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_xor_.md deleted file mode 100644 index c4c6128bfcb..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.logical_xor_.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.Tensor.logical_xor_ -### [torch.Tensor.logical_xor_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_xor_.html) -```python -torch.Tensor.logical_xor_(other) -``` - -### [paddle.Tensor.logical_xor_]() -```python -paddle.Tensor.logical_xor_(y) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| other | y | 表示输入的 Tensor ，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.nansum.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.nansum.md index c762358b3d6..275360d4a39 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.nansum.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.nansum.md @@ -3,6 +3,7 @@ ```python torch.Tensor.nansum(dim=None, keepdim=False, + *, dtype=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.std.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.std.md index 4db90f74756..5ea0d0ca2c4 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.std.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.std.md @@ -1,7 +1,7 @@ ## [ 仅参数名不一致 ]torch.Tensor.std ### [torch.Tensor.std](https://pytorch.org/docs/stable/generated/torch.Tensor.std.html?highlight=torch+tensor+std#torch.Tensor.std) ```python -torch.Tensor.std(dim=None, *, correction=1, keepdim=False) +torch.Tensor.std(dim=None, unbiased=True, keepdim=False, *, correction=1) ``` ### [paddle.Tensor.std](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor_cn.html#std-axis-none-unbiased-true-keepdim-false-name-none) @@ -16,5 +16,6 @@ paddle.Tensor.std(axis=None, unbiased=True, keepdim=False, name=None) | PyTorch | PaddlePaddle | 备注 | | ---------- | ------------ | ------- | | dim | axis | 指定对 Tensor 进行计算的轴，仅参数名不一致。 | +| unbiased | unbiased | torch v1.x 中 correction 参数名称，表示是否为无偏估计，从 v2.0 开始弃用。| | correction | unbiased | 是否使用无偏估计来计算标准差。PyTorch 支持 int 类型，Paddle 支持 bool/int 类型。仅参数名不一致。| | keepdim | keepdim | 是否在输出 Tensor 中保留减小的维度。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.var.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.var.md index adaef373036..b86d9a37856 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.var.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.Tensor.var.md @@ -1,7 +1,7 @@ ## [ 仅参数名不一致 ]torch.Tensor.var ### [torch.Tensor.var](https://pytorch.org/docs/stable/generated/torch.Tensor.var.html#torch.Tensor.var) ```python -torch.Tensor.var(dim, unbiased=True, keepdim=False) +torch.Tensor.var(dim, unbiased=True, keepdim=False, *, correction=1) ``` ### [paddle.Tensor.var](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor_cn.html#var-axis-none-unbiased-true-keepdim-false-name-none) @@ -14,6 +14,7 @@ paddle.Tensor.var(axis=None, unbiased=True, keepdim=False, name=None) | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| dim | axis | 指定对 Tensor 进行计算的轴 ，仅参数名不一致。 | -| unbiased | unbiased | 表示是否使用无偏估计来计算方差。 | -| keepdim | keepdim | 表示是否保留计算后的维度。 | +| dim | axis | 指定对 Tensor 进行计算的轴，仅参数名不一致。 | +| unbiased | unbiased | torch v1.x 中 correction 对应参数，表示是否为无偏估计，从 v2.0 开始弃用。| +| correction | unbiased | 是否使用无偏估计来计算方差。PyTorch 支持 int 类型，Paddle 支持 bool/int 类型。仅参数名不一致。| +| keepdim | keepdim | 是否在输出 Tensor 中保留减小的维度。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.add_module.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.add_module.md deleted file mode 100644 index f2f5ef7a601..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.add_module.md +++ /dev/null @@ -1,19 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.add_module -### [torch.nn.Module.add_module](https://pytorch.org/docs/stable/generated/torch.nn.Module.html?highlight=torch+nn+module+add_module#torch.nn.Module.add_module) -```python -torch.nn.Module.add_module(name, module) -``` - -### [paddle.nn.Layer.add_sublayer](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#add-sublayer-name-sublayer) -```python -paddle.nn.Layer.add_sublayer(name, sublayer) -``` - -两者功能一致，仅参数名不一致。 - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ----- | ----- | ------------------------------------------ | -| name | name | 表示子层名。 | -| module | sublayer | 表示要添加到模块中的子模块，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.buffers.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.buffers.md deleted file mode 100644 index c983df89d04..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.buffers.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.buffers -### [torch.nn.Module.buffers](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.buffers) -```python -torch.nn.Module.buffers(recurse=True) -``` - -### [paddle.nn.Layer.buffers](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#buffers-include-sublayers-true) -```python -paddle.nn.Layer.buffers(include_sublayers=True) -``` - -两者功能一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| recurse | include_sublayers | 表示是否返回子层的 buffers ，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_buffers.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_buffers.md deleted file mode 100644 index a112c895abd..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_buffers.md +++ /dev/null @@ -1,20 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.named_buffers -### [torch.nn.Module.named_buffers](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.named_buffers) -```python -torch.nn.Module.named_buffers(prefix='', recurse=True, remove_duplicate=True) -``` - -### [paddle.nn.Layer.named_buffers](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#named-buffers-prefix-include-sublayers-true) -```python -paddle.nn.Layer.named_buffers(prefix='', include_sublayers=True, remove_duplicate=True) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------------- | ------------ | ------------------------------------------------------------- | -| prefix | prefix | 在所有参数名称前加的前缀。 | -| recurse | include_sublayers | 生成该模块和所有子模块的缓冲区，仅参数名不一致。 | -| remove_duplicate | remove_duplicate | 是否删除结果中重复的模块实例。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_parameters.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_parameters.md deleted file mode 100644 index 024d7847daa..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.named_parameters.md +++ /dev/null @@ -1,20 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.named_parameters -### [torch.nn.Module.named_parameters](https://pytorch.org/docs/stable/generated/torch.nn.Module.html?highlight=torch+nn+module+named_parameters#torch.nn.Module.named_parameters) -```python -torch.nn.Module.named_parameters(prefix='', recurse=True, remove_duplicate=True) -``` - -### [paddle.nn.Layer.named_parameters](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#named-parameters-prefix-include-sublayers-true) -```python -paddle.nn.Layer.named_parameters(prefix='', include_sublayers=True, remove_duplicate=True) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------------- | ------------ | ------------------------------------------------------------- | -| prefix | prefix | 在所有参数名称前加的前缀。 | -| recurse | include_sublayers | 生成该模块和所有子模块的参数, 仅参数名不一致。 | -| remove_duplicate | remove_duplicate | 是否删除结果中的重复参数。| diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.parameters.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.parameters.md deleted file mode 100644 index 27f7b233193..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.parameters.md +++ /dev/null @@ -1,17 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.parameters -### [torch.nn.Module.parameters](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.parameters) -```python -torch.nn.Module.parameters(recurse=True) -``` - -### [paddle.nn.Layer.parameters](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#parameters-include-sublayers-true) -```python -paddle.nn.Layer.parameters(include_sublayers=True) -``` -两者功能一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| recurse | include_sublayers | 是否返回子层的参数，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_buffer.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_buffer.md deleted file mode 100644 index 0fcd61c515a..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_buffer.md +++ /dev/null @@ -1,19 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.register_buffer -### [torch.nn.Module.register_buffer](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.register_buffer) -```python -torch.nn.Module.register_buffer(name, tensor, persistent=True) -``` - -### [paddle.nn.Layer.register_buffer](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#register-buffer-name-tensor-persistable-true) -```python -paddle.nn.Layer.register_buffer(name, tensor, persistable=True) -``` -两者功能一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| name | name | 注册 buffer 的名字。可以通过此名字来访问已注册的 buffer。 | -| tensor | tensor | 将被注册为 buffer 的变量。 | -| persistent | persistable | 注册的 buffer 是否需要可持久性地保存到 state_dict 中。仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_module.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_module.md deleted file mode 100644 index 8cf866f8381..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_module.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.register_module -### [torch.nn.Module.register_module](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.register_module) -```python -torch.nn.Module.register_module(name, module) -``` - -### [paddle.nn.Layer.add_sublayer](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#add-sublayer-name-sublayer) -```python -paddle.nn.Layer.add_sublayer(name, sublayer) -``` -两者功能一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| name | name | 注册 buffer 的名字。可以通过此名字来访问已注册的 buffer。 | -| module | sublayer | 将被注册为 buffer 的模块。仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_parameter.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_parameter.md deleted file mode 100644 index fcaed2eedb0..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Module.register_parameter.md +++ /dev/null @@ -1,19 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Module.register_parameter -### [torch.nn.Module.register_parameter](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.register_parameter) -```python -torch.nn.Module.register_parameter(name, param) -``` - -### [paddle.nn.Layer.add_parameter](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#add-parameter-name-parameter) -```python -paddle.nn.Layer.add_parameter(name, parameter) -``` - -两者功能一致，仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | -------------------------------- | -| name | name | 参数名。 | -| param | parameter | Parameter 实例，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleDict.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleDict.md deleted file mode 100644 index 69dde0d7462..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleDict.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.ModuleDict -### [torch.nn.ModuleDict](https://pytorch.org/docs/stable/generated/torch.nn.ModuleDict.html?highlight=torch+nn+moduledict#torch.nn.ModuleDict) -```python -torch.nn.ModuleDict(modules=None) -``` - -### [paddle.nn.LayerDict](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/LayerDict_cn.html) -```python -paddle.nn.LayerDict(sublayers=None) -``` - -两者功能一致，参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ----- | ---------- | ---------------------------------------------------------- | -| modules | sublayers | 键值对的可迭代对象，值的类型为 paddle.nn.Layer ，参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleList.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleList.md deleted file mode 100644 index 68d26bd81d4..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.ModuleList.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.ModuleList -### [torch.nn.ModuleList](https://pytorch.org/docs/stable/generated/torch.nn.ModuleList.html?highlight=torch+nn+modulelist#torch.nn.ModuleList) -```python -torch.nn.ModuleList(modules=None) -``` - -### [paddle.nn.LayerList](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/LayerList_cn.html) -```python -paddle.nn.LayerList(sublayers=None) -``` - -两者功能一致，参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ----- | ---------- | -------------------------- | -| modules | sublayers | 要保存的子层，参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Softmax.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Softmax.md deleted file mode 100644 index 9c538e45ad2..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.Softmax.md +++ /dev/null @@ -1,18 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.Softmax -### [torch.nn.Softmax](https://pytorch.org/docs/stable/generated/torch.nn.Softmax.html?highlight=nn+softmax#torch.nn.Softmax) -```python -torch.nn.Softmax(dim=None) -``` - -### [paddle.nn.Softmax](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Softmax_cn.html#softmax) -```python -paddle.nn.Softmax(axis=- 1, - name=None) -``` - -两者功能一致且参数用法一致，仅参数名不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| dim | axis | 指定对输入 Tensor 进行运算的轴，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.functional.unfold.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.functional.unfold.md deleted file mode 100644 index d09999a3a05..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.functional.unfold.md +++ /dev/null @@ -1,22 +0,0 @@ -## [ 仅参数名不一致 ]torch.nn.functional.unfold -### [torch.nn.functional.unfold](https://pytorch.org/docs/stable/generated/torch.nn.functional.unfold.html#torch.nn.functional.unfold) -```python -torch.nn.functional.unfold(input, kernel_size, dilation=1, padding=0, stride=1) -``` - -### [paddle.nn.functional.unfold](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/functional/unfold_cn.html) -```python -paddle.nn.functional.unfold(x, kernel_sizes, strides=1, paddings=0, dilations=1, name=None) -``` - -其中功能一致, 仅参数名不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ----------- | ------------ | --------------------------------- | -| input | x | 输入 4-D Tensor，仅参数名不一致。 | -| kernel_size | kernel_sizes | 卷积核的尺寸。 | -| dilation | dilations | 卷积膨胀。 | -| padding | paddings | 每个维度的扩展，仅参数名不一致。 | -| stride | strides | 卷积步长，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.utils.parametrizations.spectral_norm.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.utils.parametrizations.spectral_norm.md new file mode 100644 index 00000000000..b28b5dd9155 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.utils.parametrizations.spectral_norm.md @@ -0,0 +1,25 @@ +## [ 仅参数名不一致 ]torch.nn.utils.parametrizations.spectral_norm + +### [torch.nn.utils.parametrizations.spectral_norm](https://pytorch.org/docs/stable/generated/torch.nn.utils.parametrizations.spectral_norm.html#torch.nn.utils.parametrizations.spectral_norm) + +```python +torch.nn.utils.parametrizations.spectral_norm(module, name='weight', n_power_iterations=1, eps=1e-12, dim=None) +``` + +### [paddle.nn.utils.spectral_norm](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/utils/spectral_norm_cn.html#spectral-norm) + +```python +paddle.nn.utils.spectral_norm(layer, name='weight', n_power_iterations=1, eps=1e-12, dim=None) +``` + +PyTorch 和 Paddle 功能一致，仅参数名不一致，具体如下： + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +| ------------------ | ------------------ | ---------------------------------------- | +| module | layer | 要添加权重谱归一化的层，仅参数名不一致。 | +| name | name | 权重参数的名字。 | +| n_power_iterations | n_power_iterations | 将用于计算的 SpectralNorm 幂迭代次数。 | +| eps | eps | 用于保证计算中的数值稳定性。 | +| dim | dim | 重塑为矩阵之前应排列到第一个的维度索引。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.sparse.addmm.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.sparse.addmm.md index ce36af87813..ffdff785ece 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.sparse.addmm.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.sparse.addmm.md @@ -2,7 +2,7 @@ ### [torch.sparse.addmm](https://pytorch.org/docs/stable/generated/torch.sparse.addmm.html?highlight=addmm#torch.sparse.addmm) ```python # PyTorch 文档有误，测试 PyTorch 第一个参数名为 input -torch.sparse.addmm(input, mat1, mat2, beta=1.0, alpha=1.0) +torch.sparse.addmm(input, mat1, mat2, *, beta=1.0, alpha=1.0) ``` ### [paddle.sparse.addmm](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/sparse/addmm_cn.html) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.LogitsProcessor.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.LogitsProcessor.md new file mode 100644 index 00000000000..799498ea9cc --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.LogitsProcessor.md @@ -0,0 +1,19 @@ +## [ 仅参数名不一致 ]transformers.LogitsProcessor +### [transformers.LogitsProcessor](https://hf-mirror.com/docs/transformers/v4.42.0/en/internal/generation_utils#logitsprocessor) +```python +transformers.LogitsProcessor(input_ids: torch.LongTensor, scores: torch.FloatTensor) +``` + +### [paddleformers.generation.LogitsProcessor](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/generation/logits_process.py#L26) +```python +paddleformers.generation.LogitsProcessor(input_ids: paddle.Tensor, scores: paddle.Tensor) +``` + +两者功能一致但参数名不一致，部分参数名不同，具体如下： + +### 参数映射 + +| transformers | PaddlePaddle | 备注 | +| ----------------- | ----------------- | --------------------------------------- | +| input_ids | input_ids | 输入(tokens) 的 id 组成的 Tensor。 | +| scores | logits | 得分组成的 Tensor，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.generation.LogitsProcessor.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.generation.LogitsProcessor.md deleted file mode 100644 index 7b8de43b763..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/transformers.generation.LogitsProcessor.md +++ /dev/null @@ -1,19 +0,0 @@ -## [ 仅参数名不一致 ]transformers.generation.LogitsProcessor -### [transformers.generation.LogitsProcessor](https://hf-mirror.com/docs/transformers/v4.42.0/en/internal/generation_utils#logitsprocessor) -```python -transformers.generation.LogitsProcessor(input_ids: torch.LongTensor, scores: torch.FloatTensor) -``` - -### [paddlenlp.generation.LogitsProcessor](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/generation/logits_process.py#L26) -```python -paddlenlp.generation.LogitsProcessor(input_ids: paddle.Tensor, scores: paddle.Tensor) -``` - -两者功能一致但参数名不一致，部分参数名不同，具体如 - -### 参数映射 - -| transformers | PaddlePaddle | 备注 | -| ----------------- | ----------------- | --------------------------------------- | -| input_ids | input_ids | 输入(tokens) 的 id 组成的 Tensor。 | -| scores | logits | 得分组成的 Tensor，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.Tensor.resize_as_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.Tensor.resize_as_.md index 81b16b1c802..2cff6719969 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.Tensor.resize_as_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.Tensor.resize_as_.md @@ -2,7 +2,7 @@ ### [torch.Tensor.resize_as_](https://pytorch.org/docs/stable/generated/torch.Tensor.resize_as_.html?highlight=resize_as#torch.Tensor.resize_as_) ```python # PyTorch 文档有误，测试第一个参数为 the_template -torch.Tensor.resize_as_(the_template, memory_format=torch.contiguous_format) +torch.Tensor.resize_as_(the_template, *, memory_format=torch.contiguous_format) ``` Paddle 无此 API，需要组合实现。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.set_num_interop_threads.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.set_num_interop_threads.md index 68dc746b956..b7689174072 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.set_num_interop_threads.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.set_num_interop_threads.md @@ -1,7 +1,7 @@ ## [ 组合替代实现 ]torch.set_num_interop_threads ### [torch.set_num_interop_threads](https://pytorch.org/docs/stable/generated/torch.set_num_interop_threads.html) ```python -torch.set_num_interop_threads() +torch.set_num_interop_threads(nthreads) ``` 设置 CPU 上用于操作间并行的线程数 (例如，在 JIT 解释器中)，Paddle 无此 API，需要组合实现。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.std_mean.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.std_mean.md index 29a497f60bc..928087eb484 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.std_mean.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.std_mean.md @@ -1,13 +1,12 @@ ## [ 组合替代实现 ]torch.std_mean ### [torch.std_mean](https://pytorch.org/docs/stable/generated/torch.std_mean.html?highlight=std_mean#torch.std_mean) ```python -# 用法一： -torch.std_mean(input, unbiased=True) -# 用法二： -torch.std_mean(input, dim, unbiased=True, keepdim=False) +torch.std_mean(input, dim=None, unbiased=True, keepdim=False, *, correction=None) ``` 用于实现返回 Tensor 的标准差和均值，PaddlePaddle 目前暂无对应 API，可使用如下代码组合实现该 API。 +> 注：torch 旧版本额外重载了 torch.std_mean(input, unbiased=True) 的签名，该用法未提供转写示例 + ### 转写示例 ```python # PyTorch 写法 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.default_collate_err_msg_format.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.default_collate_err_msg_format.md new file mode 100644 index 00000000000..1e7582465c3 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.default_collate_err_msg_format.md @@ -0,0 +1,16 @@ +## [ 组合替代实现 ]torch.utils.data.\_utils.collate.default\_collate\_err\_msg\_format +### [torch.utils.data.\_utils.collate.default\_collate\_err\_msg\_format](https://github.com/pytorch/pytorch/blob/2f023bf7b962e69c0de01b89af197388d9df27cc/torch/utils/data/_utils/collate.py#L112) +```python +torch.utils.data._utils.collate.default_collate_err_msg_format +``` + +此常量定义了当 DataLoader 无法自动将某个数据类型打成 Batch 时，抛出的那个 TypeError 的报错信息的模版。Paddle 无此 API，可替换为常量。 + +### 转写示例 +```python +# PyTorch 写法 +torch.utils.data._utils.collate.default_collate_err_msg_format.format("type") + +# Paddle 写法 +'default_collate: batch must contain tensors, numpy arrays, numbers, dicts or lists; found {}'.format("type") +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.np_str_obj_array_pattern.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.np_str_obj_array_pattern.md new file mode 100644 index 00000000000..c581e1b7bf1 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.utils.data._utils.collate.np_str_obj_array_pattern.md @@ -0,0 +1,16 @@ +## [ 组合替代实现 ]torch.utils.data.\_utils.collate.np\_str\_obj\_array\_pattern +### [torch.utils.data.\_utils.collate.np\_str\_obj\_array\_pattern](https://github.com/pytorch/pytorch/blob/2f023bf7b962e69c0de01b89af197388d9df27cc/torch/utils/data/_utils/collate.py#L20) +```python +torch.utils.data._utils.collate.np_str_obj_array_pattern +``` + +此常量定义了一个正则表达式（RegEx），用来判断一个 NumPy 数组的元素类型（dtype）是否是“字符串”或“对象”类型。 + +### 转写示例 +```python +# PyTorch 写法 +pattern = torch.utils.data._utils.collate.np_str_obj_array_pattern + +# Paddle 写法 +pattern = re.compile(r'[SaUO]') +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.var_mean.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.var_mean.md index c69e2759ef8..78f5b47b129 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.var_mean.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/torch.var_mean.md @@ -1,17 +1,12 @@ ## [ 组合替代实现 ]torch.var_mean ### [torch.var_mean](https://pytorch.org/docs/stable/generated/torch.var_mean.html?highlight=var_mean#torch.var_mean) ```python -# 用法一： -torch.var_mean(input, - unbiased=True) -# 用法二： -torch.var_mean(input, - dim, - keepdim=False, - unbiased=True) +torch.var_mean(input, dim=None, unbiased=True, keepdim=False, *, correction=None) ``` 用于实现返回 Tensor 的方差和均值，PaddlePaddle 目前暂无对应 API，可使用如下代码组合实现该 API。 +> 注：torch 旧版本额外重载了 torch.var_mean(input, unbiased=True) 的签名，该用法未提供转写示例 + ### 转写示例 ```python # PyTorch 写法 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/transformers.PreTrainedModel.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/transformers.PreTrainedModel.md index 4b7bf6b826e..af018159393 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/transformers.PreTrainedModel.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/composite_implement/transformers.PreTrainedModel.md @@ -4,7 +4,7 @@ transformers.PreTrainedModel(config: PretrainedConfig, *inputs, **kwargs) ``` -### [paddlenlp.transformers.PreTrainedModel](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/transformers/model_utils.py#L843) +### [paddleformers.transformers.PretrainedModel](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/transformers/model_utils.py#L843) ```python -paddlenlp.transformers.PreTrainedModel(*args, **kwargs) +paddleformers.transformers.PretrainedModel(*args, **kwargs) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.nn.Module.type.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.nn.Module.type.md deleted file mode 100644 index bdc7b53448e..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.nn.Module.type.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ 输入参数类型不一致 ]torch.nn.Module.type -### [torch.nn.Module.type](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.type) -```python -torch.nn.Module.type(dst_type) -``` - -### [paddle.nn.Layer.astype](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#astype-dtype-none) -```python -paddle.nn.Layer.astype(dtype=None) -``` - -两者参数用法不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------- | ------------ | --------------------------------------------------------------------------------------- | -| dst_type | dtype | PyTorch 为字符串或 PyTorch 数据类型，Paddle 为 字符串或 Paddle 数据类型，需要转写。 | - -### 转写示例 -#### dst_type 参数：数据类型 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.type(dst_type=torch.float32) - -# Paddle 写法: -module = paddle.nn.Layer() -module.astype(dtype=paddle.float32) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.alexnet.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.alexnet.md index 926d1b2bfb6..6ac99662593 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.alexnet.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.alexnet.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.alexnet ### [torchvision.models.alexnet](https://pytorch.org/vision/stable/models/generated/torchvision.models.alexnet.html) ```python -torchvision.models.alexnet(*, weights: Optional[AlexNet_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.alexnet(pretrained: bool = False, progress: bool = True, *, weights: Optional[AlexNet_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.alexnet](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/alexnet_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.alexnet(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 AlexNet_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet121.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet121.md index fcfe3f5cf5e..929f1ae1075 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet121.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet121.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.densenet121 ### [torchvision.models.densenet121](https://pytorch.org/vision/main/models/generated/torchvision.models.densenet121.html) ```python -torchvision.models.densenet121(*, weights: Optional[DenseNet121_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.densenet121(pretrained: bool = False, progress: bool = True, *, weights: Optional[DenseNet121_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.densenet121](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/densenet121_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.densenet121(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 DenseNet121_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | | ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet161.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet161.md index ac0d0c290f0..3e433fb334a 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet161.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet161.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.densenet161 ### [torchvision.models.densenet161](https://pytorch.org/vision/main/models/generated/torchvision.models.densenet161.html) ```python -torchvision.models.densenet161(*, weights: Optional[DenseNet161_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.densenet161(pretrained: bool=False, progress: bool = True, *, weights: Optional[DenseNet161_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.densenet161](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/densenet161_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.densenet161(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 DenseNet161_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet169.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet169.md index 4435932171d..2b680471149 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet169.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet169.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.densenet169 ### [torchvision.models.densenet169](https://pytorch.org/vision/main/models/generated/torchvision.models.densenet169.html) ```python -torchvision.models.densenet169(*, weights: Optional[DenseNet169_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.densenet169(pretrained: bool = False, progress: bool = True, *, weights: Optional[DenseNet169_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.densenet169](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/densenet169_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.densenet169(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 DenseNet169_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet201.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet201.md index c9e96d6af68..94ccacd8d7d 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet201.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.densenet201.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.densenet201 ### [torchvision.models.densenet201](https://pytorch.org/vision/main/models/generated/torchvision.models.densenet201.html) ```python -torchvision.models.densenet201(*, weights: Optional[DenseNet201_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.densenet201(pretrained: bool = False, progress: bool = True, *, weights: Optional[DenseNet201_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.densenet201](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/densenet201_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.densenet201(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 DenseNet201_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.googlenet.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.googlenet.md index 6915c76c35f..aa488e893da 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.googlenet.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.googlenet.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.googlenet ### [torchvision.models.googlenet](https://pytorch.org/vision/main/models/generated/torchvision.models.googlenet.html) ```python -torchvision.models.googlenet(*, weights: Optional[GoogLeNet_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.googlenet(pretrained: bool = False, progress: bool = True, *, weights: Optional[GoogLeNet_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.googlenet](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/googlenet_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.googlenet(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 GoogLeNet_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.inception_v3.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.inception_v3.md index 800d75028e3..df326a40d16 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.inception_v3.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.inception_v3.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.inception_v3 ### [torchvision.models.inception_v3](https://pytorch.org/vision/main/models/generated/torchvision.models.inception_v3.html) ```python -torchvision.models.inception_v3(*, weights: Optional[Inception_V3_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.inception_v3(pretrained: bool = False, progress: bool = True, *, weights: Optional[Inception_V3_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.inception_v3](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/inception_v3_cn.html) @@ -15,9 +15,10 @@ paddle.vision.models.inception_v3(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | +| pretrained | pretrained | 是否使用预训练权重，PyTorch 已弃用此参数。 | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 Inception_V3_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v2.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v2.md index 6a6c9fd3cde..c1d651787f7 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v2.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v2.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.mobilenet_v2 ### [torchvision.models.mobilenet_v2](https://pytorch.org/vision/stable/models/generated/torchvision.models.mobilenet_v2.html) ```python -torchvision.models.mobilenet_v2(*, weights: Optional[MobileNet_V2_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.mobilenet_v2(pretrained: bool = False, progress: bool = True, *, weights: Optional[MobileNet_V2_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.mobilenet_v2](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/mobilenet_v2_cn.html) @@ -17,9 +17,10 @@ paddle.vision.models.mobilenet_v2(pretrained=False, scale=1.0, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 MobileNet_V2_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | - | scale | 模型通道数的缩放比例，PyTorch 无此参数，Paddle 保持默认即可。 | | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_large.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_large.md index 74a78c54d7f..8547536fea9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_large.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_large.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.mobilenet_v3_large ### [torchvision.models.mobilenet_v3_large](https://pytorch.org/vision/main/models/generated/torchvision.models.mobilenet_v3_large.html) ```python -torchvision.models.mobilenet_v3_large(*, weights: Optional[MobileNet_V3_Large_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.mobilenet_v3_large(pretrained: bool = False, progress: bool = True, *, weights: Optional[MobileNet_V3_Large_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.mobilenet_v3_large](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/mobilenet_v3_large_cn.html) @@ -17,9 +17,10 @@ paddle.vision.models.mobilenet_v3_large(pretrained=False, scale=1.0, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 MobileNet_V3_Large_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | scale | 通道数缩放比例，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_small.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_small.md index 6c0ebfb1c3a..6cdc61452ef 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_small.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.mobilenet_v3_small.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.mobilenet_v3_small ### [torchvision.models.mobilenet_v3_small](https://pytorch.org/vision/main/models/generated/torchvision.models.mobilenet_v3_small.html) ```python -torchvision.models.mobilenet_v3_small(*, weights: Optional[MobileNet_V3_Small_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.mobilenet_v3_small(pretrained: bool = False, progress: bool = True, *, weights: Optional[MobileNet_V3_Small_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.mobilenet_v3_small](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/mobilenet_v3_small_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.mobilenet_v3_small(pretrained=False, scale=1.0, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 MobileNet_V3_Small_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | scale | 通道数缩放比例，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet101.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet101.md index 0c2c5052abb..480243189e1 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet101.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet101.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnet101 ### [torchvision.models.resnet101](https://pytorch.org/vision/stable/models/generated/torchvision.models.resnet101.html) ```python -torchvision.models.resnet101(*, weights: Optional[ResNet101_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnet101(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNet101_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnet101](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnet101_cn.html) @@ -17,8 +17,9 @@ paddle.vision.models.resnet101(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNet101_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet152.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet152.md index 5f15a626996..225d770eee2 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet152.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet152.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnet152 ### [torchvision.models.resnet152](https://pytorch.org/vision/stable/models/generated/torchvision.models.resnet152.html) ```python -torchvision.models.resnet152(*, weights: Optional[ResNet152_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnet152(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNet152_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnet152](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnet152_cn.html) @@ -17,8 +17,9 @@ paddle.vision.models.resnet152(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNet152_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet18.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet18.md index 0ae93c50a1d..4065ad35243 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet18.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet18.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnet18 ### [torchvision.models.resnet18](https://pytorch.org/vision/stable/models/generated/torchvision.models.resnet18.html) ```python -torchvision.models.resnet18(*, weights: Optional[ResNet18_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnet18(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNet18_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnet18](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnet18_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.resnet18(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNet18_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet34.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet34.md index d4362eb10b8..8eabd6163ed 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet34.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet34.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnet34 ### [torchvision.models.resnet34](https://pytorch.org/vision/stable/models/generated/torchvision.models.resnet34.html) ```python -torchvision.models.resnet34(*, weights: Optional[ResNet34_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnet34(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNet34_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnet34](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnet34_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.resnet34(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNet34_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet50.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet50.md index ffaa37ba076..6e89e1813fa 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet50.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnet50.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnet50 ### [torchvision.models.resnet50](https://pytorch.org/vision/stable/models/generated/torchvision.models.resnet50.html) ```python -torchvision.models.resnet50(*, weights: Optional[ResNet50_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnet50(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNet50_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnet50](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnet50_cn.html) @@ -17,8 +17,9 @@ paddle.vision.models.resnet50(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNet50_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext101_64x4d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext101_64x4d.md index 573b71ab9de..58cfe5d2b65 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext101_64x4d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext101_64x4d.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnext101_64x4d ### [torchvision.models.resnext101_64x4d](https://pytorch.org/vision/main/models/generated/torchvision.models.resnext101_64x4d.html) ```python -torchvision.models.resnext101_64x4d(*, weights: Optional[ResNeXt101_64X4D_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnext101_64x4d(pretrained: bool = False, progress: bool = True, *, weights: Optional[ResNeXt101_64X4D_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.resnext101_64x4d](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnext101_64x4d_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.resnext101_64x4d(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNeXt101_64X4D_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext50_32x4d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext50_32x4d.md index b195b1ddb37..9f33f9fc7c6 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext50_32x4d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.resnext50_32x4d.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.resnext50_32x4d ### [torchvision.models.resnext50_32x4d](https://pytorch.org/vision/main/models/generated/torchvision.models.resnext50_32x4d.html) ```python -torchvision.models.resnext50_32x4d(*, weights: Optional[ResNeXt50_32X4D_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.resnext50_32x4d(pretrained: bool=False, progress: bool = True, *, weights: Optional[ResNeXt50_32X4D_Weights] = None, progress: bool = True, **kwargs: Any) ``` ### [paddle.vision.models.resnext50_32x4d](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/resnext50_32x4d_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.resnext50_32x4d(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | -| weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNeXt50_32X4D_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ResNeXt50_32X4D_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x0_5.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x0_5.md index 81ce24bb4e9..7faf777fc91 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x0_5.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x0_5.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.shufflenet_v2_x0_5 ### [torchvision.models.shufflenet_v2_x0_5](https://pytorch.org/vision/main/models/generated/torchvision.models.shufflenet_v2_x0_5.html) ```python -torchvision.models.shufflenet_v2_x0_5(*, weights: Optional[ShuffleNet_V2_X0_5_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.shufflenet_v2_x0_5(pretrained: bool = False, progress: bool = True, *, weights: Optional[ShuffleNet_V2_X0_5_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.shufflenet_v2_x0_5](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/shufflenet_v2_x0_5_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.shufflenet_v2_x0_5(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ShuffleNet_V2_X0_5_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_0.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_0.md index 0efb4fadb4f..0a4d2c229da 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_0.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_0.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.shufflenet_v2_x1_0 ### [torchvision.models.shufflenet_v2_x1_0](https://pytorch.org/vision/main/models/generated/torchvision.models.shufflenet_v2_x1_0.html) ```python -torchvision.models.shufflenet_v2_x1_0(*, weights: Optional[ShuffleNet_V2_X1_0_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.shufflenet_v2_x1_0(pretrained: bool = False, progress: bool = True, *, weights: Optional[ShuffleNet_V2_X1_0_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.shufflenet_v2_x1_0](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/shufflenet_v2_x1_0_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.shufflenet_v2_x1_0(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ShuffleNet_V2_X1_0_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_5.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_5.md index ad2c1319887..71b1d84165d 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_5.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x1_5.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.shufflenet_v2_x1_5 ### [torchvision.models.shufflenet_v2_x1_5](https://pytorch.org/vision/main/models/generated/torchvision.models.shufflenet_v2_x1_5.html) ```python -torchvision.models.shufflenet_v2_x1_5(*, weights: Optional[ShuffleNet_V2_X1_5_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.shufflenet_v2_x1_5(pretrained: bool = False, progress: bool = True, *, weights: Optional[ShuffleNet_V2_X1_5_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.shufflenet_v2_x1_5](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/shufflenet_v2_x1_5_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.shufflenet_v2_x1_5(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ShuffleNet_V2_X1_5_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x2_0.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x2_0.md index 4b967759575..5b2079774f3 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x2_0.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.shufflenet_v2_x2_0.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.shufflenet_v2_x2_0 ### [torchvision.models.shufflenet_v2_x2_0](https://pytorch.org/vision/main/models/generated/torchvision.models.shufflenet_v2_x2_0.html) ```python -torchvision.models.shufflenet_v2_x2_0(*, weights: Optional[ShuffleNet_V2_X2_0_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.shufflenet_v2_x2_0(pretrained: bool = False, progress: bool = True, *, weights: Optional[ShuffleNet_V2_X2_0_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.shufflenet_v2_x2_0](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/shufflenet_v2_x2_0_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.shufflenet_v2_x2_0(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 ShuffleNet_V2_X2_0_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_0.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_0.md index 8ec7b2882bf..ea84b060850 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_0.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_0.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.squeezenet1_0 ### [torchvision.models.squeezenet1_0](https://pytorch.org/vision/main/models/generated/torchvision.models.squeezenet1_0.html) ```python -torchvision.models.squeezenet1_0(*, weights: Optional[SqueezeNet1_0_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.squeezenet1_0(pretrained: bool = False, progress: bool = True, *, weights: Optional[SqueezeNet1_0_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.squeezenet1_0](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/squeezenet1_0_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.squeezenet1_0(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 SqueezeNet1_0_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_1.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_1.md index bcbba04bd68..1d2744c225f 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_1.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.squeezenet1_1.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.squeezenet1_1 ### [torchvision.models.squeezenet1_1](https://pytorch.org/vision/main/models/generated/torchvision.models.squeezenet1_1.html) ```python -torchvision.models.squeezenet1_1(*, weights: Optional[SqueezeNet1_1_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.squeezenet1_1(pretrained: bool = False, progress: bool = True, *, weights: Optional[SqueezeNet1_1_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.squeezenet1_1](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/squeezenet1_1_cn.html) @@ -16,8 +16,9 @@ paddle.vision.models.squeezenet1_1(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 SqueezeNet1_1_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11.md index 07c26c0603d..36fe3003b2f 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg11 ### [torchvision.models.vgg11](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg11.html) ```python -torchvision.models.vgg11(*, weights: Optional[VGG11_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg11(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG11_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg11](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg11_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg11(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG11_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11_bn.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11_bn.md index 46d1da0e4c5..4a66eaae148 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11_bn.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg11_bn.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg11_bn ### [torchvision.models.vgg11_bn](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg11_bn.html) ```python -torchvision.models.vgg11_bn(*, weights: Optional[VGG11_BN_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg11_bn(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG11_BN_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg11](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg11_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg11(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG11_BN_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 应设置为 True。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13.md index b2ce7943e70..74731972db0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg13 ### [torchvision.models.vgg13](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg13.html) ```python -torchvision.models.vgg13(*, weights: Optional[VGG13_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg13(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG13_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg13](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg13_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg13(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG13_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13_bn.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13_bn.md index 6500ec2ecf4..6e2c3796e33 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13_bn.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg13_bn.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg13_bn ### [torchvision.models.vgg13_bn](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg13_bn.html) ```python -torchvision.models.vgg13_bn(*, weights: Optional[VGG13_BN_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg13_bn(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG13_BN_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg13](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg13_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg13(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG13_BN_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 应设置为 True。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16.md index 7fa57814499..115c1cc3f9d 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg16 ### [torchvision.models.vgg16](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg16.html) ```python -torchvision.models.vgg16(*, weights: Optional[VGG16_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg16(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG16_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg16](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg16_cn.html) @@ -17,9 +17,10 @@ paddle.vision.models.vgg16(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG16_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16_bn.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16_bn.md index b4960284034..9802ab9e703 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16_bn.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg16_bn.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg16_bn ### [torchvision.models.vgg16_bn](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg16_bn.html) ```python -torchvision.models.vgg16_bn(*, weights: Optional[VGG16_BN_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg16_bn(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG16_BN_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg16](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg16_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg16(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG16_BN_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 应设置为 True。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19.md index 5c79a81d784..ab07b86d564 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg19 ### [torchvision.models.vgg19](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg19.html) ```python -torchvision.models.vgg19(*, weights: Optional[VGG19_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg19(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG19_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg19](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg19_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg19(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG19_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 保持默认即可。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19_bn.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19_bn.md index d8524ccd497..5fe42e77003 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19_bn.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.vgg19_bn.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.vgg19_bn ### [torchvision.models.vgg19_bn](https://pytorch.org/vision/main/models/generated/torchvision.models.vgg19_bn.html) ```python -torchvision.models.vgg19_bn(*, weights: Optional[VGG19_BN_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.vgg19_bn(pretrained: bool = False, progress: bool = True, *, weights: Optional[VGG19_BN_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.vgg19](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/vgg19_cn.html) @@ -16,9 +16,10 @@ paddle.vision.models.vgg19(pretrained=False, batch_norm=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 VGG19_BN_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | batch_norm | 是否使用批归一化，PyTorch 无此参数，Paddle 应设置为 True。 | -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet101_2.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet101_2.md index 588e5b89191..e0784a33268 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet101_2.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet101_2.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.wide_resnet101_2 ### [torchvision.models.wide_resnet101_2](https://pytorch.org/vision/stable/models/generated/torchvision.models.wide_resnet101_2.html) ```python -torchvision.models.wide_resnet101_2(*, weights: Optional[WideResNet101_2_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.wide_resnet101_2(pretrained: bool = False, progress: bool = True, *, weights: Optional[WideResNet101_2_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.wide_resnet101_2](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/wide_resnet101_2_cn.html) @@ -17,8 +17,9 @@ paddle.vision.models.wide_resnet101_2(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 WideResNet101_2_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet50_2.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet50_2.md index 5686a9964e3..bedd22f89e3 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet50_2.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torchvision.models.wide_resnet50_2.md @@ -1,7 +1,7 @@ ## [ 输入参数类型不一致 ]torchvision.models.wide_resnet50_2 ### [torchvision.models.wide_resnet50_2](https://pytorch.org/vision/stable/models/generated/torchvision.models.wide_resnet50_2.html) ```python -torchvision.models.wide_resnet50_2(*, weights: Optional[WideResNet50_2_Weights] = None, progress: bool = True, **kwargs: Any) +torchvision.models.wide_resnet50_2(pretrained: bool = False, progress: bool = True, *, weights: Optional[WideResNet50_2_Weights] = None, **kwargs: Any) ``` ### [paddle.vision.models.wide_resnet50_2](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/vision/models/wide_resnet50_2_cn.html) @@ -18,8 +18,9 @@ paddle.vision.models.wide_resnet50_2(pretrained=False, **kwargs) | torchvision | PaddlePaddle | 备注 | | ----------- | ------------ | ---- | | weights | pretrained | 预训练权重，PyTorch 参数 weights 为 WideResNet50_2_Weights 枚举类或 String 类型，Paddle 参数 pretrained 为 bool 类型，需要转写。| +| pretrained | pretrained | 是否加载预训练权重。torchvision 在 0.13+ 弃用此参数。| | progress | - | 是否显示下载进度条，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| kwargs | kwargs | 附加的关键字参数。| +| **kwargs | **kwargs | 附加的关键字参数。| ### 转写示例 #### weights: 预训练权重 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_1d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_1d.md index 4e5b09564b0..af8f624edbf 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_1d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_1d.md @@ -15,12 +15,12 @@ PyTorch 与 Paddle 参数不一致，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| tensors | inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | +| *tensors | *inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | PyTorch 与 Paddle 功能一致，但对于由多个 Tensor 组成 tuple|list 输入的处理方式略有不同，具体请看转写示例。 ### 转写示例 -#### tensors: 输入为 tuple(Tensor)时 +#### *tensors: 输入为 tuple(Tensor)时 ```python # PyTorch 写法 torch.atleast_1d((x, y)) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_2d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_2d.md index f87c63f1410..1aebd282b42 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_2d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_2d.md @@ -15,12 +15,12 @@ PyTorch 与 Paddle 参数不一致，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| tensors | inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | +| *tensors | *inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | PyTorch 与 Paddle 功能一致，但对于由多个 Tensor 组成 tuple|list 输入的处理方式略有不同，具体请看转写示例。 ### 转写示例 -#### tensors: 输入为 tuple(Tensor)时 +#### *tensors: 输入为 tuple(Tensor)时 ```python # PyTorch 写法 torch.atleast_2d((x, y)) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_3d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_3d.md index 716ec961f66..b8cd4bacf41 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_3d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.atleast_3d.md @@ -15,12 +15,12 @@ PyTorch 与 Paddle 参数不一致，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| tensors | inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | +| *tensors | *inputs | 输入的 Tensor，仅当 torch 输入为 tuple(Tensor)时，两者处理方式不一致，需要转写。其他情形下均一致。 | PyTorch 与 Paddle 功能一致，但对于由多个 Tensor 组成 tuple|list 输入的处理方式略有不同，具体请看转写示例。 ### 转写示例 -#### tensors: 输入为 tuple(Tensor)时 +#### *tensors: 输入为 tuple(Tensor)时 ```python # PyTorch 写法 torch.atleast_3d((x, y)) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.distributed.optim.DistributedOptimizer.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.distributed.optim.DistributedOptimizer.md index eb8841a317f..7945f8509ad 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.distributed.optim.DistributedOptimizer.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.distributed.optim.DistributedOptimizer.md @@ -17,6 +17,6 @@ paddle.distributed.fleet.distributed_optimizer(optimizer, strategy=None) | --------------- | ------------ | --------------------------------------------------------------------- | | optimizer_class | optimizer | 优化器，PyTorch 输入为 class, Paddle 输入为实例化的 optimizer。 | | params_rref | - | 远程引用（ RRef ）列表，为 RRef 类型，代表要优化的参数。而 Paddle 在实例化 optimizer 时传入，为 Tensor 类型。 | -| args | - | 优化器实例化参数， Paddle 在实例化 optimizer 时传入。 | -| kwargs | - | 优化器实例化字典参数， Paddle 在实例化 optimizer 时传入。 | +| *args | - | 优化器实例化参数， Paddle 在实例化 optimizer 时传入。 | +| **kwargs | - | 优化器实例化字典参数， Paddle 在实例化 optimizer 时传入。 | | - | strategy | 用于切分优化器状态，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool1d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool1d.md deleted file mode 100644 index 9f6b51076e1..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool1d.md +++ /dev/null @@ -1,41 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.AvgPool1d -### [torch.nn.AvgPool1d](https://pytorch.org/docs/stable/generated/torch.nn.AvgPool1d.html?highlight=avgpool1d#torch.nn.AvgPool1d) -```python -torch.nn.AvgPool1d(kernel_size, - stride=None, - padding=0, - ceil_mode=False, - count_include_pad=True) -``` - -### [paddle.nn.AvgPool1D](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/AvgPool1D_cn.html#avgpool1d) -```python -paddle.nn.AvgPool1D(kernel_size, - stride=None, - padding=0, - exclusive=True, - ceil_mode=False, - name=None) -``` - -其中 PyTorch 的 count_include_pad 与 Paddle 的 exclusive 用法不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| kernel_size | kernel_size | 表示池化核的尺寸大小 。 | -| stride | stride | 表示步长 。 | -| padding | padding | 表示填充大小 。 | -| ceil_mode | ceil_mode | 表示是否用 ceil 函数计算输出的 height 和 width 。 | -| count_include_pad | - | 是否使用额外 padding 的值计算平均池化结果，默认为 True 。需要转写。 | -| - | exclusive | 是否不使用额外 padding 的值计算平均池化结果，默认为 True 。 | - -### 转写示例 -#### count_include_pad：是否使用额外 padding 的值计算平均池化结果 -```python -# PyTorch 写法 -torch.nn.AvgPool1D(kernel_size=2, stride=2, count_include_pad=True) - -# Paddle 写法 -paddle.nn.AvgPool1D(kernel_size=2, stride=2, exclusive=False) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool2d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool2d.md deleted file mode 100644 index ae5e0421ffd..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool2d.md +++ /dev/null @@ -1,46 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.AvgPool2d -### [torch.nn.AvgPool2d](https://pytorch.org/docs/stable/generated/torch.nn.AvgPool2d.html?highlight=avgpool2d#torch.nn.AvgPool2d) -```python -torch.nn.AvgPool2d(kernel_size, - stride=None, - padding=0, - ceil_mode=False, - count_include_pad=True, - divisor_override=None) -``` - -### [paddle.nn.AvgPool2D](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/AvgPool2D_cn.html#avgpool2d) -```python -paddle.nn.AvgPool2D(kernel_size, - stride=None, - padding=0, - ceil_mode=False, - exclusive=True, - divisor_override=None, - data_format='NCHW', - name=None) -``` - -其中 PyTorch 的 count_include_pad 与 Paddle 的 exclusive 用法不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| kernel_size | kernel_size | 表示池化核的尺寸大小 。 | -| stride | stride | 表示步长 。 | -| padding | padding | 表示填充大小 。 | -| ceil_mode | ceil_mode | 表示是否用 ceil 函数计算输出的 height 和 width 。 | -| count_include_pad | - | 是否使用额外 padding 的值计算平均池化结果，默认为 True 。 Paddle 无此参数，需要转写。 | -| divisor_override | divisor_override | 如果指定，它将用作除数，否则根据 kernel_size 计算除数。默认 None 。 | -| - | exclusive | 是否不使用额外 padding 的值计算平均池化结果，默认为 True。 | -| - | data_format | 输入和输出的数据格式， PyTorch 无此参数， Paddle 保持默认即可。 | - -### 转写示例 -#### count_include_pad：是否使用额外 padding 的值计算平均池化结果 -```python -# PyTorch 写法 -torch.nn.AvgPool2D(kernel_size=2, stride=2, count_include_pad=True) - -# Paddle 写法 -paddle.nn.AvgPool2D(kernel_size=2, stride=2, exclusive=False) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool3d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool3d.md deleted file mode 100644 index e648fee0bc0..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.AvgPool3d.md +++ /dev/null @@ -1,46 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.AvgPool3d -### [torch.nn.AvgPool3d](https://pytorch.org/docs/stable/generated/torch.nn.AvgPool3d.html?highlight=avgpool3d#torch.nn.AvgPool3d) -```python -torch.nn.AvgPool3d(kernel_size, - stride=None, - padding=0, - ceil_mode=False, - count_include_pad=True, - divisor_override=None) -``` - -### [paddle.nn.AvgPool3D](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/AvgPool3D_cn.html#avgpool3d) -```python -paddle.nn.AvgPool3D(kernel_size, - stride=None, - padding=0, - ceil_mode=False, - exclusive=True, - divisor_override=None, - data_format='NCDHW', - name=None) -``` - -其中 PyTorch 的 count_include_pad 与 Paddle 的 exclusive 用法不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| kernel_size | kernel_size | 表示池化核的尺寸大小 。 | -| stride | stride | 表示步长 。 | -| padding | padding | 表示填充大小 。 | -| ceil_mode | ceil_mode | 表示是否用 ceil 函数计算输出的 height 和 width 。 | -| count_include_pad | - | 是否使用额外 padding 的值计算平均池化结果，默认为 True 。 Paddle 无此参数，需要转写。 | -| divisor_override | divisor_override | 如果指定，它将用作除数，否则根据 kernel_size 计算除数。默认 None 。 | -| - | exclusive | 是否不使用额外 padding 的值计算平均池化结果，默认为 True。 | -| - | data_format | 输入和输出的数据格式，PyTorch 无此参数，Paddle 保持默认即可。 | - -### 转写示例 -#### count_include_pad：是否使用额外 padding 的值计算平均池化结果 -```python -# PyTorch 写法 -torch.nn.AvgPool3D(kernel_size=2, stride=2, count_include_pad=True) - -# Paddle 写法 -paddle.nn.AvgPool3D(kernel_size=2, stride=2, exclusive=False) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose1d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose1d.md index 3756df3de06..d933c9c12d0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose1d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose1d.md @@ -42,13 +42,12 @@ paddle.nn.Conv1DTranspose(in_channels, | padding | padding | 表示填充大小。 | | output_padding | output_padding | 表示输出 Tensor 额外添加的大小。 | | groups | groups | 表示分组数。 | -| bias | - | 是否在输出中添加可学习的 bias。 需要转写。 | +| bias | bias_attr | 是否在输出中添加可学习的 bias。 需要转写。 | | dilation | dilation | 表示空洞大小。 | | padding_mode | - | 填充模式，Paddle 无此参数，暂无转写方式。 | | device | - | 指定 Tensor 的设备，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | dtype | - | Tensor 的所需数据类型，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | weight_attr | 指定权重参数属性的对象，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | bias_attr | 指定偏置参数属性的对象，当`bias_attr`设置为 bool 类型与 PyTorch 的作用一致。 | | - | data_format | 输入和输出的数据格式，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose2d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose2d.md index e9d76667ee6..2c1bf5e01f9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose2d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose2d.md @@ -42,13 +42,12 @@ paddle.nn.Conv2DTranspose(in_channels, | padding | padding | 表示填充大小。 | | output_padding | output_padding | 表示输出 Tensor 额外添加的大小。 | | groups | groups | 表示分组数。 | -| bias | - | 是否在输出中添加可学习的 bias。 需要转写。 | +| bias | bias_attr | 是否在输出中添加可学习的 bias。 需要转写。 | | dilation | dilation | 表示空洞大小。 | | padding_mode | - | 填充模式，Paddle 无此参数，暂无转写方式。 | | device | - | 指定 Tensor 的设备，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | dtype | - | Tensor 的所需数据类型，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | weight_attr | 指定权重参数属性的对象，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | bias_attr | 指定偏置参数属性的对象，当`bias_attr`设置为 bool 类型与 PyTorch 的作用一致。 | | - | data_format | 输入和输出的数据格式，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose3d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose3d.md index e73a0ebd17c..81f8ca4b20f 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose3d.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.ConvTranspose3d.md @@ -44,13 +44,12 @@ paddle.nn.Conv3DTranspose(in_channels, | padding | padding | 表示填充大小。 | | output_padding | output_padding | 表示输出 Tensor 额外添加的大小。 | | groups | groups | 表示分组数。 | -| bias | - | 是否在输出中添加可学习的 bias。需要转写。 | +| bias | bias_attr | 是否在输出中添加可学习的 bias。需要转写。 | | dilation | dilation | 表示空洞大小。 | | padding_mode | - | 填充模式，Paddle 无此参数，暂无转写方式。 | | device | - | 指定 Tensor 的设备，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | dtype | - | Tensor 的所需数据类型，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | weight_attr | 指定权重参数属性的对象，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | bias_attr | 指定偏置参数属性的对象，当`bias_attr`设置为 bool 类型与 PyTorch 的作用一致。 | | - | data_format | 输入和输出的数据格式，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Linear.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Linear.md deleted file mode 100644 index ea110298bff..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Linear.md +++ /dev/null @@ -1,48 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.Linear -### [torch.nn.Linear](https://pytorch.org/docs/stable/generated/torch.nn.Linear.html?highlight=linear#torch.nn.Linear) -```python -torch.nn.Linear(in_features, - out_features, - bias=True, - device=None, - dtype=None) -``` - -### [paddle.nn.Linear](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Linear_cn.html#linear) -```python -paddle.nn.Linear(in_features, - out_features, - weight_attr=None, - bias_attr=None, - name=None) -``` - -其中 PyTorch 的 `bias` 与 Paddle 的 `bias_attr` 用法不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| in_features | in_features | 表示线性变换层输入单元的数目。 | -| out_features | out_features | 表示线性变换层输出单元的数目。 | -| bias | - | 是否在输出中添加可学习的 bias。 需要转写。 | -| device | - | 指定 Tensor 的设备，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| dtype | - | Tensor 的所需数据类型，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| - | weight_attr | 指定权重参数的属性，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | bias_attr | 指定偏置参数的属性, 当`bias_attr`设置为 bool 类型时与 PyTorch 的作用一致。 | - -### 转写示例 -#### bias: 是否在输出中添加可学习的 bias -```python -# PyTorch 写法 -torch.nn.Linear(2, 4, bias=True) - -# Paddle 写法 -paddle.nn.Linear(2, 4) -``` -```python -# PyTorch 写法 -torch.nn.Linear(2, 4, bias=False) - -# Paddle 写法 -paddle.nn.Linear(2, 4, bias_attr=False) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.xpu.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.xpu.md deleted file mode 100644 index 5f2737f10df..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.xpu.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.Module.xpu -### [torch.nn.Module.xpu](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.xpu) -```python -torch.nn.Module.xpu(device=None) -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(device="xpu") -``` - -两者参数用法不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ---------------------------------------------------------- | -| device | device | PyTorch 为设备编号，Paddle 为 xpu:设备编号，需要转写。 | - -### 转写示例 -#### device 参数：设备 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.xpu(0) - -# Paddle 写法: -module = paddle.nn.Layer() -module.to("xpu:0") -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.zero_grad.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.zero_grad.md deleted file mode 100644 index 363ea745eb7..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.Module.zero_grad.md +++ /dev/null @@ -1,28 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.Module.zero_grad -### [torch.nn.Module.zero_grad](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.zero_grad) -```python -torch.nn.Module.zero_grad(set_to_none=True) -``` - -### [paddle.nn.Layer.clear_gradients](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#clear-gradients) -```python -paddle.nn.Layer.clear_gradients(set_to_zero=True) -``` - -PyTorch 的 `Module.zero_grad` 参数与 Paddle 的 `Layer.clear_gradients` 参数用法刚好相反，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ----------- | ------------ | ------------------------------------------------ | -| set_to_none | set_to_zero | 设置如何清空梯度，PyTorch 默认 set_to_none 为 True，Paddle 默认 set_to_zero 为 True，需要转写，两者功能刚好相反，Paddle 需设置为 False。 | - -### 转写示例 -#### set_to_none 参数 -```python -# PyTorch 写法 -torch.nn.Module.zero_grad(set_to_none=True) - -# Paddle 写法 -paddle.nn.Layer.clear_gradients(set_to_zero=False) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.functional.linear.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.functional.linear.md deleted file mode 100644 index 4bbcf9fb0cb..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.nn.functional.linear.md +++ /dev/null @@ -1,34 +0,0 @@ -## [ 输入参数用法不一致 ]torch.nn.functional.linear -### [torch.nn.functional.linear](https://pytorch.org/docs/stable/generated/torch.nn.functional.linear.html?highlight=linear#torch.nn.functional.linear) -```python -torch.nn.functional.linear(input, - weight, - bias=None) -``` - -### [paddle.nn.functional.linear](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/functional/linear_cn.html) -```python -paddle.nn.functional.linear(x, - weight, - bias=None, - name=None) -``` - -两者功能一致，`weight` 参数的 `shape` 不一致，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| input | x | 表示输入的 Tensor ，仅参数名不一致。 | -| weight | weight | 表示权重 Tensor， PyTorch weight 参数的 shape 为 (out_features,in_features)，Paddle weight 参数的 shape 为 (in_features, out_features)，两者互为转置，需要转写。 | -| bias | bias | 表示偏重 Tensor 。 | - -### 转写示例 -#### weight -```python -# PyTorch 写法 -result = F.linear(x, weight=weight) - -# Paddle 写法 -result = F.linear(x, weight=weight.T) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_rank.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_rank.md new file mode 100644 index 00000000000..69f4cff8c7b --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_rank.md @@ -0,0 +1,23 @@ +## [ 仅 API 调用方式不一致 ]fairscale.nn.model_parallel.initialize.get_model_parallel_rank +### [fairscale.nn.model_parallel.initialize.get_model_parallel_rank](https://github.com/facebookresearch/fairscale/blob/164cc0f3170b4a3951dd84dda29c3e1504ac4d6e/fairscale/nn/model_parallel/initialize.py#L155) +```python +fairscale.nn.model_parallel.initialize.get_model_parallel_rank() +``` + +### [paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP.get_model_parallel_rank](https://github.com/PaddlePaddle/Paddle/blob/ddac1b431483ddc0f1ee600e799aa31fc0a75961/python/paddle/distributed/fleet/base/topology.py#L463) +```python +paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP.get_model_parallel_rank() +``` + +两者功能一致，均无参数。 + + +### 转写示例 + +```python +# PyTorch 写法 +rank = fairscale.nn.model_parallel.initialize.get_model_parallel_rank() + +# Paddle 写法 +rank = paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP.get_model_parallel_rank() +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_world_size.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_world_size.md new file mode 100644 index 00000000000..7d9329e76ed --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/fairscale.nn.model_parallel.initialize.get_model_parallel_world_size.md @@ -0,0 +1,21 @@ +## [ 仅 API 调用方式不一致 ]fairscale.nn.model_parallel.initialize.get_model_parallel_world_size +### [fairscale.nn.model_parallel.initialize.get_model_parallel_world_size](https://github.com/facebookresearch/fairscale/blob/164cc0f3170b4a3951dd84dda29c3e1504ac4d6e/fairscale/nn/model_parallel/initialize.py#L150) +```python +fairscale.nn.model_parallel.initialize.get_model_parallel_world_size() +``` + +### [paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP._mp_degree](https://github.com/PaddlePaddle/Paddle/blob/ddac1b431483ddc0f1ee600e799aa31fc0a75961/python/paddle/distributed/fleet/base/topology.py#L185) +```python +paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP._mp_degree +``` +两者功能一致，均无参数。 + +### 转写示例 + +```python +# PyTorch 写法 +world_size = fairscale.nn.model_parallel.initialize.get_model_parallel_world_size() + +# Paddle 写法 +world_size = paddle.distributed.fleet.base.topology._HYBRID_PARALLEL_GROUP._mp_degree +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/flash_attn.__version__.split.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/flash_attn.__version__.split.md new file mode 100644 index 00000000000..ebb0210ccfb --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/flash_attn.__version__.split.md @@ -0,0 +1,20 @@ +## [ 仅 API 调用方式不一致 ]flash_attn._\_version__.split +### [flash_attn._\_version__.split](https://github.com/Dao-AILab/flash-attention/blob/72e27c6320555a37a83338178caa25a388e46121/flash_attn/__init__.py) +```python +flash_attn.__version__.split(sep=None, maxsplit=-1) +``` + +### [paddle._\_version__.split](https://github.com/PaddlePaddle/Paddle/tree/develop) +```python +paddle.__version__.split(sep=None, maxsplit=-1) +``` + +### 转写示例 + +```python +# PyTorch 写法 +version = flash_attn.__version__ + +# Paddle 写法 +version = paddle.__version__ +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.bernoulli.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.bernoulli.md index aa971fbb396..9ee103ebcd2 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.bernoulli.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.bernoulli.md @@ -3,7 +3,7 @@ ### [torch.Tensor.bernoulli](https://pytorch.org/docs/stable/tensors.html#torch.Tensor.bernoulli) ```python -torch.Tensor.bernoulli(generator=None) +torch.Tensor.bernoulli(p=0.5, *, generator=None) ``` ### [paddle.bernoulli](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/bernoulli_cn.html#paddle/bernoulli_cn#cn-api-paddle-bernoulli) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.equal.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.equal.md new file mode 100644 index 00000000000..71e28e37f7e --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.equal.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.equal + +### [torch.equal](https://docs.pytorch.org/docs/stable/generated/torch.equal.html) + +```python +torch.equal(input, other) +``` + +### [paddle.compat.equal](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/__init__.py#L70) + +```python +paddle.compat.equal(input, other) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +result = torch.equal(a, b) + +# Paddle 写法 +result = paddle.compat.equal(a, b) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.max.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.max.md index 384ef3efaf4..e24cd96453b 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.max.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.max.md @@ -12,7 +12,7 @@ torch.max(*args, **kwargs) paddle.compat.max(*args, **kwargs) ``` -torch.max 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.median.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.median.md index a0ce0e54988..a6c91dc81d2 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.median.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.median.md @@ -12,7 +12,7 @@ torch.median(*args, **kwargs) paddle.compat.median(*args, **kwargs) ``` -torch.median 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.min.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.min.md index c1a3b9307d5..bc465bf0b1b 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.min.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.min.md @@ -12,7 +12,7 @@ torch.min(*args, **kwargs) paddle.compat.min(*args, **kwargs) ``` -torch.min 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nanmedian.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nanmedian.md index 5f05cb92349..4c1644a388f 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nanmedian.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nanmedian.md @@ -12,7 +12,7 @@ torch.nanmedian(*args, **kwargs) paddle.compat.nanmedian(*args, **kwargs) ``` -torch.namedian 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool1d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool1d.md new file mode 100644 index 00000000000..fc7899dcc86 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool1d.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.AvgPool1d + +### [torch.nn.AvgPool1d](https://docs.pytorch.org/docs/stable/generated/torch.nn.AvgPool1d.html) + +```python +torch.nn.AvgPool1d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True) +``` + +### [paddle.compat.nn.AvgPool1d](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/__init__.py#L67) + +```python +paddle.compat.nn.AvgPool1d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.AvgPool1d(2) + +# Paddle 写法 +layer = paddle.compat.nn.AvgPool1d(2) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool2d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool2d.md new file mode 100644 index 00000000000..475ebe2d8f3 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool2d.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.AvgPool2d + +### [torch.nn.AvgPool2d](https://docs.pytorch.org/docs/stable/generated/torch.nn.AvgPool2d.html) + +```python +torch.nn.AvgPool2d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True, divisor_override=None) +``` + +### [paddle.compat.nn.AvgPool2d](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/__init__.py#L171) + +```python +paddle.compat.nn.AvgPool2d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.AvgPool2d(kernel_size=3, stride=2, padding=0) + +# Paddle 写法 +layer = paddle.compat.nn.AvgPool2d(kernel_size=3, stride=2, padding=0) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool3d.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool3d.md new file mode 100644 index 00000000000..399b330dd68 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.AvgPool3d.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.AvgPool3d + +### [torch.nn.AvgPool3d](https://docs.pytorch.org/docs/stable/generated/torch.nn.AvgPool3d.html) + +```python +torch.nn.AvgPool3d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True, divisor_override=None) +``` + +### [paddle.compat.nn.AvgPool3d](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/__init__.py#L291) + +```python +paddle.compat.nn.AvgPool3d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.AvgPool3d(4, stride=None, padding=0, ceil_mode=False, count_include_pad=True) + +# Paddle 写法 +layer = paddle.compat.nn.AvgPool3d(4, stride=None, padding=0, ceil_mode=False, count_include_pad=True) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Linear.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Linear.md new file mode 100644 index 00000000000..36f617913f1 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Linear.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.Linear + +### [torch.nn.Linear](https://docs.pytorch.org/docs/stable/generated/torch.nn.Linear.html) + +```python +torch.nn.Linear(in_features, out_features, bias=True, device=None, dtype=None) +``` + +### [paddle.compat.nn.Linear](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/__init__.py#L470) + +```python +paddle.compat.nn.Linear(in_features, out_features, bias=True, device=None, dtype=None) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.Linear(in_features=16, out_features=16) + +# Paddle 写法 +layer = paddle.compat.nn.Linear(in_features=16, out_features=16) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.apply.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.apply.md deleted file mode 100644 index caa662004a4..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.apply.md +++ /dev/null @@ -1,36 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module.apply - -### [torch.nn.Module.apply](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.apply) - -```python -torch.nn.Module.apply(fn) -``` - -### [paddle.nn.Layer.apply](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/apply_cn.html#paddle/nn/Layer/apply_cn#cn-api-paddle-nn-Layer-apply) - -```python -paddle.nn.Layer.apply(fn) -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -def init_weights(m): - pass -net = torch.nn.Sequential( - torch.nn.Linear(2, 2, bias=False), torch.nn.Linear(2, 2, bias=False) -) -net.apply(init_weights) - -# Paddle 写法 -def init_weights(m): - pass -net = paddle.nn.Sequential( - paddle.nn.Linear(in_features=2, out_features=2, bias_attr=False), - paddle.nn.Linear(in_features=2, out_features=2, bias_attr=False), -) -net.apply(init_weights) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.children.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.children.md deleted file mode 100644 index 7a045d52274..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.children.md +++ /dev/null @@ -1,29 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module.children - -### [torch.nn.Module.children](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.children) - -```python -torch.nn.Module.children() -``` - -### [paddle.nn.Layer.children](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/children_cn.html#paddle/nn/Layer/children_cn#cn-api-paddle-nn-Layer-children) - -```python -paddle.nn.Layer.children() -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -model = torch.nn.Sequential(net1, net2) -for layer in model.children(): - print(layer) - -# Paddle 写法 -model = paddle.nn.Sequential(net1, net2) -for layer in model.children(): - print(layer) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.eval.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.eval.md deleted file mode 100644 index 1f3a502ef15..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.eval.md +++ /dev/null @@ -1,27 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module.eval - -### [torch.nn.Module.eval](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval) - -```python -torch.nn.Module.eval() -``` - -### [paddle.nn.Layer.eval](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/eval_cn.html#paddle/nn/Layer/eval_cn#cn-api-paddle-nn-Layer-eval) - -```python -paddle.nn.Layer.eval() -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -model = torch.nn.Linear(10, 10) -model.eval() - -# Paddle 写法 -model = paddle.nn.Linear(10, 10) -model.eval() -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.md deleted file mode 100644 index 658d27804d3..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.md +++ /dev/null @@ -1,45 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module - -### [torch.nn.Module](https://pytorch.org/docs/stable/generated/torch.nn.Module.html) - -```python -torch.nn.Module(*args, **kwargs) -``` - -### [paddle.nn.Layer](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#paddle/nn/Layer_cn#cn-api-paddle-nn-Layer) - -```python -paddle.nn.Layer(name_scope=None, dtype='float32') -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -class MLP(torch.nn.Module): - def __init__(self, input_size, hidden_size, output_size): - super(MLP, self).__init__() - self.fc1 = torch.nn.Linear(input_size, hidden_size) - with torch.no_grad(): - self.fc1.weight.fill_(1.0) - self.fc1.bias.fill_(0.1) - - def forward(self, x): - x = self.fc1(x) - return x - -# Paddle 写法 -class MLP(paddle.nn.Layer): - def __init__(self, input_size, hidden_size, output_size): - super(MLP, self).__init__() - self.fc1 = paddle.nn.Linear(in_features=input_size, out_features=hidden_size) - with paddle.no_grad(): - self.fc1.weight.fill_(1.0) - self.fc1.bias.fill_(0.1) - - def forward(self, x): - x = self.fc1(x) - return x -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.modules.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.modules.md deleted file mode 100644 index 3e0dd9bb523..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.modules.md +++ /dev/null @@ -1,29 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module.modules - -### [torch.nn.Module.modules](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.modules) - -```python -torch.nn.Module.modules() -``` - -### [paddle.nn.Layer.sublayers](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/sublayers_cn.html#paddle/nn/Layer/sublayers_cn#cn-api-paddle-nn-Layer-sublayers) - -```python -paddle.nn.Layer.sublayers(include_self=False) -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -model = torch.nn.Sequential(net1, net2) -for layer in model.modules(): - print(layer) - -# Paddle 写法 -model = paddle.nn.Sequential(net1, net2) -for layer in model.sublayers(): - print(layer) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.named_children.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.named_children.md deleted file mode 100644 index b63dbef1d97..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Module.named_children.md +++ /dev/null @@ -1,29 +0,0 @@ -## [ 仅 API 调用方式不一致 ]torch.nn.Module.named_children - -### [torch.nn.Module.named_children](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.named_children) - -```python -torch.nn.Module.named_children() -``` - -### [paddle.nn.Layer.named_children](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/named_children_cn.html#paddle/nn/Layer/named_children_cn#cn-api-paddle-nn-Layer-named_children) - -```python -paddle.nn.Layer.named_children() -``` - -两者功能一致，但调用方式不一致，具体如下： - -### 转写示例 - -```python -# PyTorch 写法 -model = torch.nn.Sequential(OrderedDict([("wfs", l), ("wfs1", l1)])) -for name, module in model.named_children(): - print(name, module) - -# Paddle 写法 -model = paddle.nn.Sequential(OrderedDict([("wfs", l), ("wfs1", l1)])) -for name, module in model.named_children(): - print(name, module) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.MultiheadAttention.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.MultiheadAttention.md new file mode 100644 index 00000000000..3e29dbaff19 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.MultiheadAttention.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.MultiheadAttention + +### [torch.nn.MultiheadAttention](https://docs.pytorch.org/docs/stable/generated/torch.nn.MultiheadAttention.html) + +```python +torch.nn.MultiheadAttention(embed_dim, num_heads, dropout=0.0, bias=True, add_bias_kv=False, add_zero_attn=False, kdim=None, vdim=None, batch_first=False, device=None, dtype=None) +``` + +### [paddle.compat.nn.MultiheadAttention](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/transformer.py#L29) +```python +paddle.compat.nn.MultiheadAttention(embed_dim, num_heads, dropout=0.0, bias=True, add_bias_kv=False, add_zero_attn=False, kdim=None, vdim=None, batch_first=False, device=None, dtype=None) + +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.MultiheadAttention(128, 8) + +# Paddle 写法 +layer = paddle.compat.nn.MultiheadAttention(128, 8) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Softmax.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Softmax.md new file mode 100644 index 00000000000..487952999a7 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Softmax.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.Softmax + +### [torch.nn.Softmax](https://docs.pytorch.org/docs/stable/generated/torch.nn.Softmax.html) + +```python +torch.nn.Softmax(dim=None) +``` + +### [paddle.compat.nn.Softmax](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/__init__.py#L613) + +```python +paddle.compat.nn.Softmax(dim=None) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +layer = torch.nn.Softmax() + +# Paddle 写法 +layer = paddle.compat.nn.Softmax() +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Unfold.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Unfold.md index 1ec08363426..7a69d77ccf0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Unfold.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Unfold.md @@ -12,7 +12,7 @@ torch.nn.Unfold(kernel_size, dilation=1, padding=0, stride=1) paddle.compat.nn.Unfold(kernel_size, dilation=1, padding=0, stride=1) ``` -两者功能一致，但调用方式不一致，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.linear.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.linear.md new file mode 100644 index 00000000000..255632853c1 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.linear.md @@ -0,0 +1,24 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.functional.linear + +### [torch.nn.functional.linear](https://docs.pytorch.org/docs/stable/generated/torch.nn.functional.linear.html) + +```python +torch.nn.functional.linear(input, weight, bias=None) +``` + +### [paddle.compat.nn.functional.linear](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/functional/__init__.py#L204) +```python +paddle.compat.nn.functional.linear(input, weight, bias=None) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +out = torch.nn.functional.linear(input, weight, bias) + +# Paddle 写法 +out = paddle.compat.nn.functional.linear(input, weight, bias) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.pad.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.pad.md index 279e4cdce82..7a1eb5afd89 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.pad.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.pad.md @@ -12,7 +12,7 @@ torch.nn.functional.pad(input, pad, mode="constant", value=None) paddle.compat.nn.functional.pad(input, pad, mode="constant", value=0.0) ``` -两者功能一致，但调用方式不一致，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.scaled_dot_product_attention.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.scaled_dot_product_attention.md new file mode 100644 index 00000000000..dbf79d394c6 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.scaled_dot_product_attention.md @@ -0,0 +1,27 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.functional.scaled_dot_product_attention + +### [torch.nn.functional.scaled_dot_product_attention](https://docs.pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) +```python +torch.nn.functional.scaled_dot_product_attention( + query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, scale=None, enable_gqa=False +) +``` + +### [paddle.compat.nn.functional.scaled_dot_product_attention](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/functional/sdpa.py#L25) +```python +paddle.compat.nn.functional.scaled_dot_product_attention( + query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, scale=None, enable_gqa=False +) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +out = torch.nn.functional.scaled_dot_product_attention(query, key, value) + +# Paddle 写法 +out = paddle.compat.nn.functional.scaled_dot_product_attention(query, key, value) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.softmax.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.softmax.md index 22a8ac882d2..58b715574fd 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.softmax.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.softmax.md @@ -12,7 +12,7 @@ torch.nn.functional.softmax(input, dim=None, _stacklevel=3, dtype=None) paddle.compat.nn.functional.softmax(input, dim=None, _stacklevel=3, dtype=None, out=None) ``` -两者功能一致，但调用方式不一致，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.unfold.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.unfold.md new file mode 100644 index 00000000000..5c9374b49cf --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.unfold.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.nn.functional.unfold + +### [torch.nn.functional.unfold](https://docs.pytorch.org/docs/stable/generated/torch.nn.functional.unfold.html) + +```python +torch.nn.functional.unfold(input, kernel_size, dilation=1, padding=0, stride=1) +``` + +### [paddle.compat.nn.functional.unfold](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/compat/nn/transformer.py#L29) +```python +paddle.compat.nn.functional.unfold(input, kernel_size, dilation=1, padding=0, stride=1) + +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +out = torch.nn.functional.unfold(input, kernel_size) + +# Paddle 写法 +out = paddle.compat.nn.functional.unfold(input, kernel_size) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.sort.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.sort.md index e3b905bcfb5..0ae429d6500 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.sort.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.sort.md @@ -12,7 +12,7 @@ torch.sort(input, dim=-1, descending=False, *, stable=False, out=None) paddle.compat.sort(input, dim=-1, descending=False, stable=False, out=None) ``` -两者功能一致，但调用方式不一致，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.split.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.split.md index 96707a62103..85b54b6bb13 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.split.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.split.md @@ -12,7 +12,7 @@ torch.split(tensor, split_size_or_sections, dim=0) paddle.compat.split(tensor, split_size_or_sections, dim=0) ``` -两者功能一致，但调用方式不一致，具体如下： +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.unique.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.unique.md new file mode 100644 index 00000000000..0c18e0a4cad --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.unique.md @@ -0,0 +1,25 @@ +## [ 仅 API 调用方式不一致 ]torch.unique + +### [torch.unique](https://docs.pytorch.org/docs/stable/generated/torch.unique.html) + +```python +torch.unique(input, sorted=True, return_inverse=False, return_counts=False, dim=None) +``` + +### [paddle.compat.unique](https://docs.pytorch.org/docs/stable/generated/torch.unique.html) + +```python +paddle.compat.unique(input, sorted=True, return_inverse=False, return_counts=False, dim=None) +``` + +两者功能和用法一致，但 API 路径不一致，只需修改 torch 前缀为 paddle.compat，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +result = torch.unique(x) + +# Paddle 写法 +result = paddle.compat.unique(x) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.RandomSampler.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.RandomSampler.md similarity index 53% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.RandomSampler.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.RandomSampler.md index 8541153a758..387901aceee 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.RandomSampler.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.RandomSampler.md @@ -1,4 +1,4 @@ -## [ 暂未分类 ]torch.utils.data.RandomSampler +## [ 仅 API 调用方式不一致 ]torch.utils.data.RandomSampler ### [torch.utils.data.RandomSampler](https://pytorch.org/docs/stable/data.html#torch.utils.data.RandomSampler) ```python torch.utils.data.RandomSampler(data_source, replacement=False, num_samples=None, generator=None) @@ -9,4 +9,14 @@ torch.utils.data.RandomSampler(data_source, replacement=False, num_samples=None, paddle.io.RandomSampler(data_source, replacement=False, num_samples=None, generator=None) ``` -两者参数完全一致，具体如下： +两者功能一致，但调用方式不一致，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +torch.utils.data.RandomSampler(data_source=[1, 2, 3, 4, 5], replacement=False, num_samples=None, generator=None) + +# Paddle 写法 +paddle.io.RandomSampler(data_source=[1, 2, 3, 4, 5], replacement=False, num_samples=None, generator=None) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.WeightedRandomSampler.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.WeightedRandomSampler.md similarity index 63% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.WeightedRandomSampler.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.WeightedRandomSampler.md index fcefd17f784..e61a969c2c2 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/torch.utils.data.WeightedRandomSampler.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.utils.data.WeightedRandomSampler.md @@ -1,4 +1,4 @@ -## [ 暂未分类 ]torch.utils.data.WeightedRandomSampler +## [ 仅 API 调用方式不一致 ]torch.utils.data.WeightedRandomSampler ### [torch.utils.data.WeightedRandomSampler](https://pytorch.org/docs/stable/data.html#torch.utils.data.WeightedRandomSampler) ```python torch.utils.data.WeightedRandomSampler(weights, @@ -14,4 +14,14 @@ paddle.io.WeightedRandomSampler(weights, replacement=True) ``` -两者参数完全一致，具体如下： +两者功能一致，但调用方式不一致，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +sampler = torch.utils.data.WeightedRandomSampler([0.5, 0.5], 20, True) + +# Paddle 写法 +sampler = paddle.io.WeightedRandomSampler([0.5, 0.5], 20, True) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.AddedToken.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.AddedToken.md new file mode 100644 index 00000000000..b39288c5743 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.AddedToken.md @@ -0,0 +1,22 @@ +## [ 仅 API 调用方式不一致 ]transformers.AddedToken +### [transformers.AddedToken](https://github.com/huggingface/transformers/blob/d625294d79341662784495551abdf45e6cb9372f/src/transformers/tokenization_utils_base.py#L84) +```python +transformers.AddedToken(content: str, single_word=False, lstrip=False, rstrip=False, special=False, normalized=None) +``` + +### [paddleformers.transformers.AddedToken](https://github.com/PaddlePaddle/PaddleFormers/blob/8a544d93fab8984c223218d2e31328ffefb01f67/paddleformers/transformers/legacy/tokenizer_utils_base.py#L84) +```python +paddleformers.transformers.AddedToken(content: str, single_word=False, lstrip=False, rstrip=False, special=False, normalized=None) +``` + +两者功能一致，但调用方式不一致，具体如下： + +### 转写示例 + +```python +# PyTorch 写法 +transformers.AddedToken("newToken", single_word=True) + +# Paddle 写法 +paddleformers.transformers.AddedToken("newToken", single_word=True) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.GenerationConfig.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.GenerationConfig.md new file mode 100644 index 00000000000..da54662be63 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.GenerationConfig.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.GenerationConfig +### [transformers.GenerationConfig](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/text_generation#transformers.GenerationConfig) +```python +transformers.GenerationConfig(**kwargs) +``` + +### [paddleformers.generation.GenerationConfig](https://github.com/PaddlePaddle/PaddleFormers/blob/8a544d93fab8984c223218d2e31328ffefb01f67/paddleformers/generation/configuration_utils.py#L62) +```python +paddleformers.generation.GenerationConfig(**kwargs) +``` + +两者功能一致，均接受可变参数 kwargs 初始化。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PreTrainedTokenizer.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PreTrainedTokenizer.md new file mode 100644 index 00000000000..5a0dd459db2 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PreTrainedTokenizer.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.PreTrainedTokenizer +### [transformers.PreTrainedTokenizer](https://huggingface.co/docs/transformers/main_classes/tokenizer#transformers.PythonBackend) +```python +transformers.PreTrainedTokenizer(**kwargs) +``` + +### [paddleformers.transformers.PretrainedTokenizer](https://github.com/PaddlePaddle/PaddleFormers/blob/8a544d93fab8984c223218d2e31328ffefb01f67/paddleformers/generation/configuration_utils.py#L62) +```python +paddleformers.transformers.PretrainedTokenizer(**kwargs) +``` + +两者功能一致，均接受可变参数 kwargs 初始化。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PretrainedConfig.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PretrainedConfig.md new file mode 100644 index 00000000000..e11264af44f --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.PretrainedConfig.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.PretrainedConfig +### [transformers.PretrainedConfig](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/configuration#transformers.PretrainedConfig) +```python +transformers.PretrainedConfig(*kwargs) +``` + +### [paddleformers.transformers.PretrainedConfig](https://github.com/PaddlePaddle/PaddleFormers/blob/8a544d93fab8984c223218d2e31328ffefb01f67/paddleformers/transformers/configuration_utils.py#L572) +```python +paddleformers.transformers.PretrainedConfig(*kwargs) +``` + +两者功能一致，均接受可变参数 kwargs 初始化。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.StoppingCriteriaList.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.StoppingCriteriaList.md new file mode 100644 index 00000000000..841ac96f1be --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.StoppingCriteriaList.md @@ -0,0 +1,35 @@ +## [ 仅 API 调用方式不一致 ]transformers.StoppingCriteriaList + +### [transformers.StoppingCriteriaList](https://github.com/huggingface/transformers/blob/d625294d79341662784495551abdf45e6cb9372f/src/transformers/generation/stopping_criteria.py#L503) + +```python +transformers.StoppingCriteriaList() + +``` + +### [paddleformers.generation.StoppingCriteriaList](https://github.com/PaddlePaddle/PaddleFormers/blob/ca66f8dd619a6b2e17fa901042277501b2ed3230/paddleformers/generation/stopping_criteria.py#L72) + +```python +paddleformers.generation.StoppingCriteriaList() + +``` + +两者功能一致，均为继承自 Python 原生 `list` 的容器类，用于管理多个停止条件。参数定义与用法完全一致。 + +### 转写示例 + +```python +# PyTorch 写法 +from transformers import StoppingCriteriaList, MaxLengthCriteria + +criteria = MaxLengthCriteria(max_length=20) +stopping_criteria = StoppingCriteriaList([criteria]) + + +# Paddle 写法 +from[paddleformers.generation import StoppingCriteriaList, MaxLengthCriteria + +criteria = MaxLengthCriteria(max_length=20) +stopping_criteria = StoppingCriteriaList([criteria]) + +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.LogitsProcessorList.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.LogitsProcessorList.md new file mode 100644 index 00000000000..397780bd462 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.LogitsProcessorList.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.LogitsProcessorList +### [transformers.LogitsProcessorList](https://hf-mirror.com/docs/transformers/v4.42.0/en/internal/generation_utils#transformers.LogitsProcessorList) +```python +transformers.LogitsProcessorList() +``` + +### [paddleformers.generation.LogitsProcessorList](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/transformers/bloom/processor.py#L24) +```python +paddleformers.generation.LogitsProcessorList() +``` + +虚基类，用于被继承。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.utils.GenerateOutput.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.utils.GenerateOutput.md new file mode 100644 index 00000000000..c997830b5f0 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.generation.utils.GenerateOutput.md @@ -0,0 +1,13 @@ +## [ 仅 API 调用方式不一致 ]transformers.generation.utils.GenerateOutput +### [transformers.generation.utils.GenerateOutput](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/output#transformers.generation.utils.GenerateOutput) +```python +transformers.generation.utils.GenerateOutput(last_hidden_state: FloatTensor = None, past_key_values: Optional = None, hidden_states: Optional = None, attentions: Optional = None) +``` + +### [paddleformers.transformers.model_outputs.BaseModelOutput](https://github.com/PaddlePaddle/PaddleFormers/blob/4bbcdc3405d529cdefee7d4e1caa7dfb13715893/paddleformers/transformers/model_outputs.py#L513) +```python +paddleformers.transformers.model_outputs.BaseModelOutput(last_hidden_state: paddle.Tensor = None, past_key_values: Optional[Tuple[Tuple[paddle.Tensor]]] = None, hidden_states: Optional[Tuple[paddle.Tensor]] = None, attentions: Optional[Tuple[paddle.Tensor]] = None) +``` + + +二者功能一致，参数完全一致，用于封装模型的输出。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.BaseModelOutputWithPast.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.BaseModelOutputWithPast.md new file mode 100644 index 00000000000..156d741ebfc --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.BaseModelOutputWithPast.md @@ -0,0 +1,13 @@ +## [ 仅 API 调用方式不一致 ]transformers.modeling_outputs.BaseModelOutputWithPast +### [transformers.modeling_outputs.BaseModelOutputWithPast](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPast) +```python +transformers.modeling_outputs.BaseModelOutputWithPast(last_hidden_state: FloatTensor = None, past_key_values: Optional = None, hidden_states: Optional = None, attentions: Optional = None) +``` + +### [paddleformers.transformers.model_outputs.BaseModelOutputWithPast](https://github.com/PaddlePaddle/PaddleFormers/blob/4bbcdc3405d529cdefee7d4e1caa7dfb13715893/paddleformers/transformers/model_outputs.py#L590) +```python +paddleformers.transformers.model_outputs.BaseModelOutputWithPast(last_hidden_state: paddle.Tensor = None, past_key_values: Optional[Tuple[Tuple[paddle.Tensor]]] = None, hidden_states: Optional[Tuple[paddle.Tensor]] = None, attentions: Optional[Tuple[paddle.Tensor]] = None) +``` + + +二者功能一致，参数完全一致，用于封装模型的输出。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.CausalLMOutputWithPast.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.CausalLMOutputWithPast.md new file mode 100644 index 00000000000..15d8d397378 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.modeling_outputs.CausalLMOutputWithPast.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.modeling_outputs.CausalLMOutputWithPast +### [transformers.modeling_outputs.CausalLMOutputWithPast](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/output#transformers.modeling_outputs.CausalLMOutputWithPast) +```python +transformers.modeling_outputs.CausalLMOutputWithPast(loss: Optional = None, logits: FloatTensor = None, past_key_values: Optional = None, hidden_states: Optional = None, attentions: Optional = None) +``` + +### [paddleformers.transformers.model_outputs.CausalLMOutputWithPast](https://github.com/PaddlePaddle/PaddleFormers/blob/4bbcdc3405d529cdefee7d4e1caa7dfb13715893/paddleformers/transformers/model_outputs.py#L922) +```python +paddleformers.transformers.model_outputs.CausalLMOutputWithPast(loss: Optional[paddle.Tensor] = None, logits: paddle.Tensor = None, past_key_values: Optional[Tuple[Tuple[paddle.Tensor]]] = None, hidden_states: Optional[Tuple[paddle.Tensor]] = None, attentions: Optional[Tuple[paddle.Tensor]] = None) +``` + +二者功能一致，参数完全一致，用于封装模型的输出。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.utils.add_start_docstrings_to_model_forward.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.utils.add_start_docstrings_to_model_forward.md new file mode 100644 index 00000000000..f18e8fcc9a7 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/transformers.utils.add_start_docstrings_to_model_forward.md @@ -0,0 +1,12 @@ +## [ 仅 API 调用方式不一致 ]transformers.utils.add_start_docstrings_to_model_forward +### [transformers.utils.add_start_docstrings_to_model_forward](https://github.com/huggingface/transformers/blob/44b231671db25974cfebcdae34402ad5099bf37a/src/transformers/utils/doc.py#L45) +```python +transformers.utils.add_start_docstrings_to_model_forward(*docstr) +``` + +### [paddleformers.trainer.utils.add_start_docstrings_to_model_forward](https://github.com/PaddlePaddle/PaddleFormers/blob/4bbcdc3405d529cdefee7d4e1caa7dfb13715893/paddleformers/trainer/utils/doc.py#L27) +```python +paddleformers.trainer.utils.add_start_docstrings_to_model_forward(*docstr) +``` + +两者功能一致，用于添加公共的前向传播文档字符串。 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/transformers.PreTrainedTokenizer.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/others/transformers.PreTrainedTokenizer.md deleted file mode 100644 index 4484c63e33a..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/others/transformers.PreTrainedTokenizer.md +++ /dev/null @@ -1,12 +0,0 @@ -## [ 暂未分类 ]transformers.PreTrainedTokenizer -### [transformers.PreTrainedTokenizer](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/tokenizer#transformers.PreTrainedTokenizer) -```python -transformers.PreTrainedTokenizer(**kwargs) -``` - -### [paddlenlp.transformers.PretrainedTokenizer](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/transformers/tokenizer_utils.py#L881) -```python -paddlenlp.transformers.PretrainedTokenizer(**kwargs) -``` - -功能一致，参数完全一致，具体如下： diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.eq_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.eq_.md similarity index 59% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.eq_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.eq_.md index d17df4da3d9..efae9ca1bd9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.eq_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.eq_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.eq_ +## [ 返回参数类型不一致 ]torch.Tensor.eq_ ### [torch.Tensor.eq_](https://pytorch.org/docs/stable/generated/torch.Tensor.eq_.html) ```python torch.Tensor.eq_(other) @@ -9,20 +9,21 @@ torch.Tensor.eq_(other) paddle.Tensor.equal_(y) ``` -其中，PyTorch 与 Paddle 的 `other` 参数所支持类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | |----------------------------------|------------------------------| ------------------------------------------------------ | | other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other：输入为 Number +#### other/返回值：输入为 Number ```python # PyTorch 写法 result = x.eq_(2) # Paddle 写法 -result = x.equal_(paddle.to_tensor(2)) -``` +dtype = x.dtype +result = x.equal_(paddle.to_tensor(2)).cast_(dtype) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ge_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ge_.md similarity index 58% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ge_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ge_.md index c38cd507f9e..382e44d77b0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ge_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ge_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.ge_ +## [ 返回参数类型不一致 ]torch.Tensor.ge_ ### [torch.Tensor.ge_](https://pytorch.org/docs/stable/ge_nerated/torch.Tensor.ge_.html) ```python torch.Tensor.ge_(other) @@ -9,20 +9,21 @@ torch.Tensor.ge_(other) paddle.Tensor.greater_equal_(y) ``` -其中，PyTorch 与 Paddle 的 `other` 参数所支持类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | |----------------------------------|------------------------------| ------------------------------------------------------ | | other | y | 输入的 Tensor ，PyTorch 支持 Tensor 和 Python Number，Paddle 仅支持 Tensor，需要转写。 | - +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other +#### other/返回值 ```python # PyTorch 写法 result = x.ge_(other=2) # Paddle 写法 -result = x.greater_equal_(y=paddle.to_tensor(2)) +dtype = x.dtype +result = x.greater_equal_(y=paddle.to_tensor(2)).cast_(dtype) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.gt_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.gt_.md similarity index 59% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.gt_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.gt_.md index 303bf9c8ac4..37ee9aed6fa 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.gt_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.gt_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.gt_ +## [ 返回参数类型不一致 ]torch.Tensor.gt_ ### [torch.Tensor.gt_](https://pytorch.org/docs/stable/generated/torch.Tensor.gt_.html) ```python torch.Tensor.gt_(other) @@ -9,19 +9,20 @@ torch.Tensor.gt_(other) paddle.Tensor.greater_than_(y) ``` -其中 Paddle 和 PyTorch 的 `other` 参数所支持的数据类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | |----------------------------------|------------------------------| ------------------------------------------------------ | | other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | - +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other：输入为 Number +#### other/返回值：输入为 Number ```python # PyTorch 写法 result = x.gt_(2) # Paddle 写法 -result = x.greater_than_(paddle.to_tensor(2)) +dtype = x.dtype +result = x.greater_than_(paddle.to_tensor(2)).cast_(dtype) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.le_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.le_.md similarity index 59% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.le_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.le_.md index d4f59899495..ff8ed3a2f52 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.le_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.le_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.le_ +## [ 返回参数类型不一致 ]torch.Tensor.le_ ### [torch.Tensor.le_](https://pytorch.org/docs/stable/generated/torch.Tensor.le_.html) ```python torch.Tensor.le_(other) @@ -9,19 +9,20 @@ torch.Tensor.le_(other) paddle.Tensor.less_equal_(y) ``` -其中 Paddle 和 PyTorch 的 `other` 参数所支持的数据类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | |----------------------------------|------------------------------| ------------------------------------------------------ | | other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | - +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other：输入为 Number +#### other/返回值：输入为 Number ```python # PyTorch 写法 result = x.le_(2) # Paddle 写法 -result = x.less_equal_(paddle.to_tensor(2)) +dtype = x.dtype +result = x.less_equal_(paddle.to_tensor(2)).cast_(dtype) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_and_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_and_.md new file mode 100644 index 00000000000..a5585f56370 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_and_.md @@ -0,0 +1,33 @@ +## [ 返回参数类型不一致 ]torch.Tensor.logical_and_ + +### [torch.Tensor.logical_and_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_and_.html) + +```python +torch.Tensor.logical_and_(other) +``` + +### [paddle.Tensor.logical_and_]() + +```python +paddle.Tensor.logical_and_(y) +``` + +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +|-----------------------------------|------------------------------| ------------------------------------------------------ | +| other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | + + +### 转写示例 +#### other/返回值 +```python +# PyTorch 写法 +x.logical_and_(tensor_y) + +# Paddle 写法 +_x_dtype_ = x.dtype +x.logical_and_(tensor_y).cast(_x_dtype_) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_not_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_not_.md new file mode 100644 index 00000000000..2159e2412fa --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_not_.md @@ -0,0 +1,33 @@ +## [ 返回参数类型不一致 ]torch.Tensor.logical_not_ + +### [torch.Tensor.logical_not_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_not_.html) + +```python +torch.Tensor.logical_not_(other) +``` + +### [paddle.Tensor.logical_not_]() + +```python +paddle.Tensor.logical_not_(y) +``` + +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +|-----------------------------------|------------------------------| ------------------------------------------------------ | +| other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | + + +### 转写示例 +#### other/返回值 +```python +# PyTorch 写法 +x.logical_not_(tensor_y) + +# Paddle 写法 +_x_dtype_ = x.dtype +x.logical_not_(tensor_y).cast(_x_dtype_) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_or_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_or_.md new file mode 100644 index 00000000000..e3eda160254 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_or_.md @@ -0,0 +1,33 @@ +## [ 返回参数类型不一致 ]torch.Tensor.logical_or_ + +### [torch.Tensor.logical_or_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_or_.html) + +```python +torch.Tensor.logical_or_(other) +``` + +### [paddle.Tensor.logical_or_]() + +```python +paddle.Tensor.logical_or_(y) +``` + +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +|-----------------------------------|------------------------------| ------------------------------------------------------ | +| other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | + + +### 转写示例 +#### other/返回值 +```python +# PyTorch 写法 +x.logical_or_(tensor_y) + +# Paddle 写法 +_x_dtype_ = x.dtype +x.logical_or_(tensor_y).cast(_x_dtype_) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_xor_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_xor_.md new file mode 100644 index 00000000000..51ec271a697 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.logical_xor_.md @@ -0,0 +1,33 @@ +## [ 返回参数类型不一致 ]torch.Tensor.logical_xor_ + +### [torch.Tensor.logical_xor_](https://pytorch.org/docs/stable/generated/torch.Tensor.logical_xor_.html) + +```python +torch.Tensor.logical_xor_(other) +``` + +### [paddle.Tensor.logical_xor_]() + +```python +paddle.Tensor.logical_xor_(y) +``` + +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +|-----------------------------------|------------------------------| ------------------------------------------------------ | +| other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | + + +### 转写示例 +#### other/返回值 +```python +# PyTorch 写法 +x.logical_xor_(tensor_y) + +# Paddle 写法 +_x_dtype_ = x.dtype +x.logical_xor_(tensor_y).cast(_x_dtype_) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.Tensor.lstsq.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lstsq.md similarity index 77% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.Tensor.lstsq.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lstsq.md index a22d2fc1c2f..0db291b78a0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_usage_diff/torch.Tensor.lstsq.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lstsq.md @@ -1,7 +1,7 @@ -## [ 输入参数用法不一致 ]torch.Tensor.lstsq +## [ 返回参数类型不一致 ]torch.Tensor.lstsq ### [torch.Tensor.lstsq](https://pytorch.org/docs/1.9.0/generated/torch.Tensor.lstsq.html?highlight=torch%20tensor%20lstsq#torch.Tensor.lstsq) ```python -torch.Tensor.lstsq(A) +torch.Tensor.lstsq(other, rcond=None, *, driver=None) ``` ### [paddle.Tensor.lstsq]() @@ -9,20 +9,18 @@ torch.Tensor.lstsq(A) paddle.Tensor.lstsq(y, rcond=None, driver=None, name=None) ``` -两者功能一致，输入参数用法不一致，具体如下： +两者功能一致，但是 Paddle 与 PyTorch 返回值类型与数量不一致，请参考参数映射： ### 参数映射 | PyTorch | PaddlePaddle | 备注 | | ------- | ------------ | ----------------------------------------------------------------------------------------------- | -| A | - | 线性方程组系数矩阵，Paddle 需要转写。 | -| - | y | 线性方程组右边的矩阵，Paddle 需要转写。 | +| other | y | 线性方程组系数矩阵，仅参数名不一致。 | | - | rcond | 用来决定 x 有效秩的 float 型浮点数。PyTorch 无此参数，Paddle 保持默认即可。 | | - | driver | 用来指定计算使用的 LAPACK 库方法。PyTorch 无此参数，Paddle 保持默认即可。 | | 返回值 | 返回值 | PyTorch 返回 solution、QR ，Paddle 返回 solution、residuals、rank、 singular_values，Paddle 与 PyTorch 仅第一个返回值相同，其他返回值结果不同，暂无转写方式。 | ### 转写示例 -#### A 参数 ```python # PyTorch 写法: A = torch.tensor([[1, 1, 1], diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.lt_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lt_.md similarity index 59% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.lt_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lt_.md index fbdfccf23f7..439494c6ab4 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.lt_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.lt_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.lt_ +## [ 返回参数类型不一致 ]torch.Tensor.lt_ ### [torch.Tensor.lt_](https://pytorch.org/docs/stable/generated/torch.Tensor.lt_.html) ```python torch.Tensor.lt_(other) @@ -9,19 +9,20 @@ torch.Tensor.lt_(other) paddle.Tensor.less_than_(y) ``` -其中 Paddle 和 PyTorch 的 `other` 参数所支持的数据类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | |----------------------------------|------------------------------| ------------------------------------------------------ | | other | y | 表示输入的 Tensor ，PyTorch 支持 Python Number 和 Tensor 类型， Paddle 仅支持 Tensor 类型。当输入为 Python Number 类型时，需要转写。 | - +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other：输入为 Number +#### other/返回值：输入为 Number ```python # PyTorch 写法 result = x.lt_(2) # Paddle 写法 -result = x.less_than_(paddle.to_tensor(2)) +dtype = x.dtype +result = x.less_than_(paddle.to_tensor(2)).cast_(dtype) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ne_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ne_.md similarity index 55% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ne_.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ne_.md index 34c74cb0f57..112278151b9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/input_args_type_diff/torch.Tensor.ne_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.ne_.md @@ -1,4 +1,4 @@ -## [ 输入参数类型不一致 ]torch.Tensor.ne_ +## [ 返回参数类型不一致 ]torch.Tensor.ne_ ### [torch.Tensor.ne_](https://pytorch.org/docs/stable/generated/torch.Tensor.ne_.html) ```python torch.Tensor.ne_(other) @@ -9,20 +9,21 @@ torch.Tensor.ne_(other) paddle.Tensor.not_equal_(y) ``` -其中，Paddle 与 PyTorch 的 `other` 参数所支持类型不一致，具体如下： +返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型。 ### 参数映射 | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ----------------------------------------------- | | other | y | 比较的元素，PyTorch 支持 Tensor 和 Python Number，Paddle 仅支持 Tensor，需要转写。 | - +| 返回值 | 返回值 | 返回 Tensor 的数据类型不一致，PyTorch 返回数据类型与输入 Tensor 一致， Paddle 返回 paddle.bool 类型，需要转写。 | ### 转写示例 -#### other：比较的元素 +#### other/返回值：比较的元素 ```python # PyTorch 写法 y = x.ne_(other=2) # Paddle 写法 -y = x.not_equal_(y=paddle.to_tensor(2)) +dtype = x.dtype +y = x.not_equal_(y=paddle.to_tensor(2)).cast_(dtype) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.sort.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.sort.md index bb80012d2e2..5c5bb8b37bd 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.sort.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.Tensor.sort.md @@ -1,7 +1,7 @@ ## [ 返回参数类型不一致 ]torch.Tensor.sort ### [torch.Tensor.sort](https://pytorch.org/docs/stable/generated/torch.Tensor.sort.html#torch-tensor-sort) ```python -torch.Tensor.sort(dim=-1, descending=False, stable=False) +torch.Tensor.sort(dim=-1, descending=False, *, stable=False) ``` ### [paddle.Tensor.sort](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/sort_cn.html#sort) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.equal.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.equal.md deleted file mode 100644 index a7eb4a41034..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/torch.equal.md +++ /dev/null @@ -1,33 +0,0 @@ -## [ 返回参数类型不一致 ]torch.equal -### [torch.equal](https://pytorch.org/docs/stable/generated/torch.equal.html?highlight=equal#torch.equal) -```python -torch.equal(input, - other) -``` - -### [paddle.equal_all](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/equal_all_cn.html#equal-all) -```python -paddle.equal_all(x, - y, - name=None) -``` - -两者功能一致但返回参数类型不同，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ----------------------------- | -| input | x | 输入 Tensor，仅参数名不一致。 | -| other | y | 输入 Tensor，仅参数名不一致。 | -| 返回值 | 返回值 | PyTorch 返回 bool 类型，Paddle 返回 0-D bool Tensor，需要转写。| - -### 转写示例 -#### 返回值 -```Python -# torch 中的写法 -torch.equal(x, y) - -# paddle 中的写法 -paddle.equal_all(x, y).item() -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedModel.generate.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedModel.generate.md index 4572c963e70..8be95ea86e9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedModel.generate.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedModel.generate.md @@ -4,9 +4,9 @@ transformers.PreTrainedModel.generate(input, **kwargs) ``` -### [paddlenlp.transformers.PreTrainedModel.generate](https://github.com/PaddlePaddle/PaddleNLP/blob/88d4b19bc6865fb28c11d2ce83d07c3b4b8dc423/paddlenlp/generation/utils.py#L604) +### [paddleformers.transformers.PretrainedModel.generate](https://github.com/PaddlePaddle/PaddleNLP/blob/88d4b19bc6865fb28c11d2ce83d07c3b4b8dc423/paddlenlp/generation/utils.py#L604) ```python -paddlenlp.transformers.PreTrainedModel.generate(input_ids, **kwargs) +paddleformers.transformers.PretrainedModel.generate(input_ids, **kwargs) ``` ### 参数映射 @@ -23,6 +23,5 @@ paddlenlp.transformers.PreTrainedModel.generate(input_ids, **kwargs) res = transformers.PreTrainedModel.generate(input = input_x) # Paddle 写法 -temp_res = paddlenlp.transformers.PreTrainedModel.generate(input_ids = input_x) +temp_res = paddleformers.transformers.PretrainedModel.generate(input_ids = input_x) res = paddle.concat((input_x,temp_res[0]),axis=-1) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedTokenizer.encode.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedTokenizer.encode.md index 47376a41d93..97a2d3bdddb 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedTokenizer.encode.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/output_args_type_diff/transformers.PreTrainedTokenizer.encode.md @@ -4,9 +4,9 @@ transformers.PreTrainedTokenizer.encode(text, **kwargs) ``` -### [paddlenlp.transformers.PreTrainedTokenizer.encode](https://github.com/PaddlePaddle/PaddleNLP/blob/88d4b19bc6865fb28c11d2ce83d07c3b4b8dc423/paddlenlp/transformers/tokenizer_utils_base.py#L2369) +### [paddleformers.transformers.PreTrainedTokenizer.encode](https://github.com/PaddlePaddle/PaddleNLP/blob/88d4b19bc6865fb28c11d2ce83d07c3b4b8dc423/paddlenlp/transformers/tokenizer_utils_base.py#L2369) ```python -paddlenlp.transformers.PreTrainedTokenizer.encode(text, **kwargs) +paddleformers.transformers.PreTrainedTokenizer.encode(text, **kwargs) ``` ### 参数映射 @@ -23,5 +23,5 @@ paddlenlp.transformers.PreTrainedTokenizer.encode(text, **kwargs) transformers.PreTrainedTokenizer.encode(text) # Paddle 写法 -paddlenlp.transformers.PreTrainedTokenizer.encode(text)["input_ids"] +paddleformers.transformers.PreTrainedTokenizer.encode(text)["input_ids"] ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.autograd.Function.backward.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.autograd.Function.backward.md index f3ac16f1aeb..1aea2e670d3 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.autograd.Function.backward.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.autograd.Function.backward.md @@ -16,5 +16,4 @@ Paddle 相比 PyTorch 支持更多其他参数，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------ | ------------ | ------------------------------------------------------------------- | | ctx | ctx | 上下文对象。 | -| grad_outputs | args | forward 输出 Tensor 的梯度，仅参数名不一致。 | -| - | kwargs | forward 输出 Tensor 的梯度，PyTorch 无此参数，Paddle 保持默认即可。 | +| *grad_outputs | *args | forward 输出 Tensor 的梯度，仅参数名不一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.cov.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.cov.md index 160936ac753..8c1b58a43f5 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.cov.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.cov.md @@ -2,6 +2,7 @@ ### [torch.cov](https://pytorch.org/docs/stable/generated/torch.cov.html?highlight=cov#torch.cov) ```python torch.cov(input, + *, correction=1, fweights=None, aweights=None) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.linalg.vander.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.linalg.vander.md index 914b0c56e9c..5e18ebcf579 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.linalg.vander.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.linalg.vander.md @@ -1,7 +1,7 @@ ## [ paddle 参数更多 ]torch.linalg.vander ### [torch.linalg.vander](https://pytorch.org/docs/stable/generated/torch.linalg.vander.html#torch.linalg.vander) ```python -torch.linalg.vander(x, N=None) +torch.linalg.vander(x, *, N=None) ``` ### [paddle.vander](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/vander_cn.html) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.bfloat16.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.bfloat16.md deleted file mode 100644 index 4a070e67079..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.bfloat16.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.bfloat16 -### [torch.nn.Module.bfloat16](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.bfloat16) -```python -torch.nn.Module.bfloat16() -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(dtype=paddle.bfloat16) -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | --------------------------------------------------------- | -| - | dtype | 转换的数据类型，Paddle 为 paddle.bfloat16，需要转写。 | - -### 转写示例 -#### dtype 参数：转换的数据类型 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.bfloat16() - -# Paddle 写法: -module = paddle.nn.Layer() -module.to(dtype=paddle.bfloat16) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cpu.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cpu.md deleted file mode 100644 index 122225b318c..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cpu.md +++ /dev/null @@ -1,20 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.cpu -### [torch.nn.Module.cpu](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.cpu) -```python -torch.nn.Module.cpu() -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(device=None, dtype=None, blocking=None) -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | --------------------------------- | -| - | device | 移动到的设备，PyTorch 无此参数，Paddle 设置为 "cpu"。 | -| - | dtype | 数据类型，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | blocking | 是否阻塞，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cuda.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cuda.md deleted file mode 100644 index fa39f78c99b..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.cuda.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.cuda -### [torch.nn.Module.cuda](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.cuda) -```python -torch.nn.Module.cuda(device=None) -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(device="gpu") -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ---------------------------------------------------------- | -| device | device | PyTorch 为设备编号，Paddle 为 gpu:设备编号，需要转写。 | - -### 转写示例 -#### device 参数：设备 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.cuda(0) - -# Paddle 写法: -module = paddle.nn.Layer() -module.to("gpu:0") -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.double.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.double.md deleted file mode 100644 index d714a8736e7..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.double.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.double -### [torch.nn.Module.double](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.double) -```python -torch.nn.Module.double() -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(dtype="float64") -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ------------------------------------------------- | -| - | dtype | 转换的数据类型，Paddle 为 float64，需要转写。 | - -### 转写示例 -#### dtype 参数：转换的数据类型 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.double() - -# Paddle 写法: -module = paddle.nn.Layer() -module.to(dtype="float64") -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.float.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.float.md deleted file mode 100644 index ae4e7219bd1..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.float.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.float -### [torch.nn.Module.float](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.float) -```python -torch.nn.Module.float() -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(dtype="float32") -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ------------------------------------------------- | -| - | dtype | 转换的数据类型，Paddle 为 float32，需要转写。 | - -### 转写示例 -#### dtype 参数：转换的数据类型 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.float() - -# Paddle 写法: -module = paddle.nn.Layer() -module.to(dtype="float32") -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.half.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.half.md deleted file mode 100644 index 166824948b8..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.half.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.half -### [torch.nn.Module.half](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.half) -```python -torch.nn.Module.half() -``` - -### [paddle.nn.Layer.to](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#to-device-none-dtype-none-blocking-none) -```python -paddle.nn.Layer.to(dtype="float16") -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------------ | ------------------------------------------------- | -| - | dtype | 转换的数据类型，Paddle 为 float16，需要转写。 | - -### 转写示例 -#### dtype 参数：转换的数据类型 -```python -# PyTorch 写法: -module = torch.nn.Module() -module.half() - -# Paddle 写法: -module = paddle.nn.Layer() -module.to(dtype="float16") -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.named_modules.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.named_modules.md deleted file mode 100644 index 3fb2e8a7ba6..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.named_modules.md +++ /dev/null @@ -1,21 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.named_modules -### [torch.nn.Module.named_modules](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.named_modules) -```python -torch.nn.Module.named_modules(memo=None, prefix='', remove_duplicate=True) -``` - -### [paddle.nn.Layer.named_sublayers](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#named-sublayers-prefix-include-self-false-layers-set-none) -```python -paddle.nn.Layer.named_sublayers(prefix='', include_self=False, layers_set=None, remove_duplicate=True) -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------------- | ------------ | ------------------------------------------------------------- | -| memo | layers_set | 用来记录已经加入结果的子层的集合，仅参数名不一致。 | -| prefix | prefix | 在所有参数名称前加的前缀。 | -| remove_duplicate | remove_duplicate | 是否删除结果中重复的模块实例。 | -| - | include_self | 是否包含该层自身，PyTorch 无此参数，Paddle 需设为 True 才与 Pytorch 一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.state_dict.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.state_dict.md deleted file mode 100644 index 95f56d190ce..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nn.Module.state_dict.md +++ /dev/null @@ -1,22 +0,0 @@ -## [ paddle 参数更多 ]torch.nn.Module.state_dict -### [torch.nn.Module.state_dict](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.state_dict) -```python -torch.nn.Module.state_dict(destination, prefix='', keep_vars=False) -``` - -### [paddle.nn.Layer.state_dict](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#state-dict-destination-none-include-sublayers-true-use-hook-true) -```python -paddle.nn.Layer.state_dict(destination=None, include_sublayers=True, structured_name_prefix='', use_hook=True, keep_vars=True) -``` - -Paddle 相比 PyTorch 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| destination | destination | 所有参数和可持久性 buffers 都将存放在 destination 中 。 | -| prefix | structured_name_prefix | 添加到参数和缓冲区名称的前缀。 | -| keep_vars | keep_vars | 状态字典中返回的 Tensor 是否与 autograd 分离，PyTorch 默认值为 False，Paddle 为 True，Paddle 需设置为与 PyTorch 一致。 | -| - | include_sublayers | 包括子层的参数和 buffers, PyTorch 无此参数，Paddle 保持默认即可。 | -| - | use_hook | 将_state_dict_hooks 中注册的函数应用于 destination, PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.nanmean.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.nanmean.md index 74b447ef569..124f05d1860 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.nanmean.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.nanmean.md @@ -3,6 +3,7 @@ ```python torch.Tensor.nanmean(dim=None, keepdim=False, + *, dtype=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.numpy.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.numpy.md index f4b60f7a7a9..7d35e68941d 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.numpy.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.numpy.md @@ -1,7 +1,7 @@ ## [ torch 参数更多 ]torch.Tensor.numpy ### [torch.Tensor.numpy](https://pytorch.org/docs/stable/generated/torch.Tensor.numpy.html?highlight=numpy#torch.Tensor.numpy) ```python -torch.Tensor.numpy(force=False) +torch.Tensor.numpy(*, force=False) ``` ### [paddle.Tensor.numpy](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor_cn.html#numpy) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.resize_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.resize_.md index 161e8f878e2..a97747850c6 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.resize_.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.resize_.md @@ -1,6 +1,6 @@ -## [ torch 参数更多 ]torch.Tensor.resize_ +## [ torch 参数更多 ]torch.Tensor.resize\_ -### [torch.Tensor.resize_](https://pytorch.org/docs/stable/generated/torch.Tensor.resize_.html) +### [torch.Tensor.resize\_](https://pytorch.org/docs/stable/generated/torch.Tensor.resize_.html) ```python torch.Tensor.resize_(*sizes, memory_format=torch.contiguous_format) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round.md deleted file mode 100644 index 36fbcb2fee0..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ torch 参数更多 ]torch.Tensor.round -### [torch.Tensor.round](https://pytorch.org/docs/stable/generated/torch.Tensor.round.html#torch.Tensor.round) -```python -torch.Tensor.round(decimals=0) -``` - -### [paddle.Tensor.round](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor_cn.html#round-name-none) -```python -paddle.Tensor.round(name=None) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------- | ------- | ------- | -| decimals | - | 舍入小数位数。需要转写。 | - -### 转写示例 -#### decimals：要舍入到的小数位数 -```python -# PyTorch 写法 -torch.tensor([3.345, 5.774]).round(decimals=2) - -# Paddle 写法 -(paddle.to_tensor([3.345, 5.774]) * 1e2).round() / 1e2 - -# 注：Paddle 可使用 10 的 decimals 次方来实现 -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round_.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round_.md deleted file mode 100644 index 2159ee53fb0..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.Tensor.round_.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ torch 参数更多 ]torch.Tensor.round_ -### [torch.Tensor.round_](https://pytorch.org/docs/stable/generated/torch.Tensor.round_.html#torch.Tensor.round_) -```python -torch.Tensor.round_(decimals=0) -``` - -### [paddle.Tensor.round_](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/Tensor_cn.html#round-name-none) -```python -paddle.Tensor.round_(name=None) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| -------- | ------- | ------- | -| decimals | - | 舍入小数位数。需要转写。 | - -### 转写示例 -#### decimals：要舍入到的小数位数 -```python -# PyTorch 写法 -torch.tensor([3.345, 5.774]).round_(decimals=2) - -# Paddle 写法 -(paddle.to_tensor([3.345, 5.774]) * 1e2).round_() / 1e2 - -# 注：Paddle 可使用 10 的 decimals 次方来实现 -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.baddbmm.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.baddbmm.md deleted file mode 100644 index 7cdffa1bcec..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.baddbmm.md +++ /dev/null @@ -1,33 +0,0 @@ -## [ torch 参数更多 ]torch.baddbmm -### [torch.baddbmm](https://pytorch.org/docs/stable/generated/torch.baddbmm.html?highlight=baddbmm#torch.baddbmm) -```python -torch.baddbmm(input, batch1, batch2, beta=1, alpha=1, out=None) -``` - -### [paddle.baddbmm](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/baddbmm_cn.html) -```python -paddle.baddbmm(input, x, y, alpha=1.0, beta=1.0, name=None) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------- | ------- | ------- | -| input | input | 表示输入的 Tensor 。 | -| batch1 | x | 表示输入的第一个 Tensor ，仅参数名不一致。 | -| batch2 | y | 表示输入的第二个 Tensor ，仅参数名不一致。 | -| beta | beta | 表示乘以 input 的标量。 | -| alpha | alpha | 表示乘以 batch1 * batch2 的标量。 | -| out | - | 表示输出的 Tensor ， Paddle 无此参数，需要转写。 | - -### 转写示例 -#### out: 输出的 Tensor -```python -# PyTorch 写法 -torch.baddbmm(input, batch1, batch2, beta, alpha, out=output) - -# Paddle 写法 -paddle.assign(paddle.baddbmm(input, x, y, beta, alpha), output) -``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.cholesky_inverse.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.cholesky_inverse.md index e7920426d39..1c3f6a43a82 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.cholesky_inverse.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.cholesky_inverse.md @@ -1,7 +1,7 @@ ## [ torch 参数更多 ]torch.cholesky_inverse ### [torch.cholesky_inverse](https://pytorch.org/docs/stable/generated/torch.cholesky_inverse.html?highlight=cholesky_inverse#torch.cholesky_inverse) ```python -torch.cholesky_inverse(input, upper=False, out=None) +torch.cholesky_inverse(input, upper=False, *, out=None) ``` ### [paddle.linalg.cholesky_inverse](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/linalg/cholesky_inverse_cn.html#cholesky-inverse) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.frexp.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.frexp.md index 36fd1e2d373..ed0e08297dc 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.frexp.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.frexp.md @@ -2,6 +2,7 @@ ### [torch.frexp](https://pytorch.org/docs/stable/generated/torch.frexp.html?highlight=frexp#torch.frexp) ```python torch.frexp(input, + *, out=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.inverse.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.inverse.md index 562b7e1d72c..dbe8477786e 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.inverse.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.inverse.md @@ -4,9 +4,9 @@ torch.inverse(input, *, out=None) ``` -### [paddle.linalg.inv](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/linalg/inv_cn.html) +### [paddle.inverse](https://github.com/PaddlePaddle/Paddle/blob/5721d267e434c18fa64ff2b99839c7cb6d4cc04d/python/paddle/tensor/math.py#L2900) ```python -paddle.linalg.inv(x, name=None) +paddle.inverse(x, name=None) ``` PyTorch 相比 Paddle 支持更多其他参数，具体如下： diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.jit.load.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.jit.load.md index e7606e72d94..ac283e72340 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.jit.load.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.jit.load.md @@ -18,7 +18,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | f | path | PyTorch 为文件对象或文件名包含后缀，Paddle 为文件名不包含后缀，读取 .pdiparams，.pdmodel 等后缀文件。需要转写。 | | map_location | - | 存储位置，Paddle 无此参数，暂无转写方式。 | | \_extra_files | - | 额外加载的文件，Paddle 无此参数，暂无转写方式。 | -| - | configs | 其他用于兼容的载入配置选项，PyTorch 无此参数，Paddle 保持默认即可。 | +| - | **configs | 其他用于兼容的载入配置选项，PyTorch 无此参数，Paddle 保持默认即可。 | ### 转写示例 #### f 参数用法不同 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.eigvals.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.eigvals.md index 5ffe4ebfe03..5dc9cc5d5c9 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.eigvals.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.eigvals.md @@ -2,6 +2,7 @@ ### [torch.linalg.eigvals](https://pytorch.org/docs/stable/generated/torch.linalg.eigvals.html?highlight=torch+linalg+eigvals#torch.linalg.eigvals) ```python torch.linalg.eigvals(input, + *, out=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.matrix_rank.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.matrix_rank.md index 2e6d33c7ca4..ab32c8f5ec7 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.matrix_rank.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.matrix_rank.md @@ -1,8 +1,7 @@ ## [ torch 参数更多 ]torch.linalg.matrix_rank ### [torch.linalg.matrix_rank](https://pytorch.org/docs/stable/generated/torch.linalg.matrix_rank.html?highlight=matrix_rank#torch.linalg.matrix_rank) ```python -torch.linalg.matrix_rank(A, *, atol=None, rtol=None ,hermitian=False, out=None) -torch.linalg.matrix_rank(x, tol=None, hermitian=False, *, name=None) +torch.linalg.matrix_rank(input, tol=None, hermitian = False, *, atol=None, rtol=None ,hermitian=False, out=None) ``` ### [paddle.linalg.matrix_rank](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/linalg/matrix_rank_cn.html) @@ -15,7 +14,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| A | x | 输入的 Tensor ，仅参数名不一致。 | +| input | x | 输入的 Tensor ，仅参数名不一致。 | | tol | tol | 阈值，类型为 Tensor 或者 float，参数完全一致。 | | hermitian | hermitian | 表示输入矩阵是否是 Hermitian 矩阵，类型为 bool ，参数完全一致。 | | atol | atol | 绝对阈值，类型为 Tensor 或者 float，参数完全一致。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.multi_dot.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.multi_dot.md index b9e3f99c25a..1e4fe21ed21 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.multi_dot.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.linalg.multi_dot.md @@ -2,6 +2,7 @@ ### [torch.linalg.multi_dot](https://pytorch.org/docs/stable/generated/torch.linalg.multi_dot.html?highlight=torch+linalg+multi_dot#torch.linalg.multi_dot) ```python torch.linalg.multi_dot(tensors, + *, out=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.load.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.load.md index 358a9da4ab2..af1b5e25291 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.load.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.load.md @@ -26,6 +26,6 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | map_location | - | 表示如何重新映射存储位置，Paddle 无此参数，暂无转写方式。 | | pickle_module | - | 表示用于 unpickling 元数据和对象的模块，Paddle 无此参数，暂无转写方式。 | | weights_only | - | 指示 unpickler 是否应限制为仅加载张量、原始类型和字典，Paddle 无此参数，暂无转写方式。 | -| pickle_load_args | - | 传递给 pickle_module.load（）和 pickle_module.Unpickler（）的可选关键字参数，Paddle 无此参数，暂无转写方式。 | +| **pickle_load_args | - | 传递给 pickle_module.load（）和 pickle_module.Unpickler（）的可选关键字参数，Paddle 无此参数，暂无转写方式。 | | mmap | - | 指示是否使用 mmap 文件，Paddle 无此参数，暂无转写方式。 | -| - | configs | 表示其他用于兼容的载入配置选项。PyTorch 无此参数，Paddle 保持默认即可。 | +| - | **configs | 表示其他用于兼容的载入配置选项。PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.meshgrid.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.meshgrid.md index 0261da5f87a..ff8207586d2 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.meshgrid.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.meshgrid.md @@ -14,7 +14,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | PyTorch | PaddlePaddle | 备注 | | ------------- | ------------ | ------------------------------------------------------ | -| tensors | args | 输入的 Tensor 列表，仅参数名不一致。 | +| *tensors | *args | 输入的 Tensor 列表，仅参数名不一致。 | | indexing | - | tensor 的组合模式。Paddle 无此参数，需要转写。 | ### 转写示例 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.multiprocessing.spawn.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.multiprocessing.spawn.md index 3c85912f80f..01acc1eaa62 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.multiprocessing.spawn.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.multiprocessing.spawn.md @@ -20,4 +20,4 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | nprocs | nprocs | 启动进程的数目。Paddle 与 PyTorch 默认值不同，Paddle 应设置为 `1`。 | | join | join | 对所有启动的进程执行阻塞的 join，等待进程执行结束。 | | start_method | - | 启动方式。 PyTorch 已弃用，Paddle 无此参数，对网络训练结果无影响，可直接删除。 | -| - | options | 其他初始化并行执行环境的配置选项。 PyTorch 无此参数， Paddle 保持默认即可。 | +| - | **options | 其他初始化并行执行环境的配置选项。 PyTorch 无此参数， Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.mv.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.mv.md index 342bf00882b..167c6fd71d0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.mv.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.mv.md @@ -1,7 +1,7 @@ ## [ torch 参数更多 ]torch.mv ### [torch.mv](https://pytorch.org/docs/stable/generated/torch.mv.html?highlight=torch+mv#torch.mv) ```python -torch.mv(input, vec, out=None) +torch.mv(input, vec, *, out=None) ``` ### [paddle.mv](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/mv_cn.html) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nanmean.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nanmean.md index 2a92b28878d..7f95eaa3421 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nanmean.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nanmean.md @@ -4,6 +4,7 @@ torch.nanmean(input, dim=None, keepdim=False, + *, dtype=None, out=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nansum.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nansum.md similarity index 62% rename from docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nansum.md rename to docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nansum.md index 63102b02584..3e295f38717 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.nansum.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nansum.md @@ -1,7 +1,7 @@ -## [ paddle 参数更多 ]torch.nansum +## [ torch 参数更多 ]torch.nansum ### [torch.nansum](https://pytorch.org/docs/stable/generated/torch.nansum.html) ```python -torch.nansum(input, *, dtype=None) +torch.nansum(input, dim=None, keepdim=False, *, dtype=None, out=None) ``` ### [paddle.nansum](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/nansum_cn.html) @@ -16,6 +16,18 @@ Paddle 比 PyTorch 支持更多参数，具体如下： | PyTorch | PaddlePaddle | 备注 | | ----------- | ------------ | ----------------------------------------------------------------------------------------- | | input | x | 输入的 Tensor，数据类型为：float16、float32、float64、int32 或 int64。仅参数名不一致。 | -| - | axis | 求和运算的维度。PyTorch 无此参数，Paddle 保持默认即可。 | +| dim | axis | 求和运算的维度。仅参数名不一致。 | +| keep_dim | keepdim | 是否在输出 Tensor 中保留减小的维度。| | dtype | dtype | 输出变量的数据类型。若参数为空，则输出变量的数据类型和输入变量相同，默认值为 None。 | -| - | keepdim | 是否在输出 Tensor 中保留减小的维度。PyTorch 无此参数，Paddle 保持默认即可。| +| out | - | 表示输出的 Tensor,可选项，Paddle 无此参数，需要转写。| + + +### 转写示例 +#### out：指定输出 +```python +# PyTorch 写法 +torch.nansum(t, dim=1，out=y) + +# Paddle 写法 +paddle.assign(paddle.nansum(t, dim=1), y) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.load_state_dict.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.load_state_dict.md deleted file mode 100644 index 626dff9cd5e..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.load_state_dict.md +++ /dev/null @@ -1,20 +0,0 @@ -## [ torch 参数更多 ]torch.nn.Module.load_state_dict -### [torch.nn.Module.load_state_dict](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.load_state_dict) -```python -torch.nn.Module.load_state_dict(state_dict, strict=True) -``` - -### [paddle.nn.Layer.set_state_dict](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#set-state-dict-state-dict-use-structured-name-true) -```python -paddle.nn.Layer.set_state_dict(state_dict, use_structured_name=True) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| state_dict | state_dict | 包含所有参数和可持久性 buffers 的 dict。 | -| strict | - | 设置所加载参数字典的 key 是否能够严格匹配，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| - | use_structured_name | 是否将使用 Layer 的结构性变量名作为 dict 的 key，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_hook.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_hook.md index f3221f7ea29..f91026e9600 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_hook.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_hook.md @@ -4,9 +4,9 @@ torch.nn.Module.register_forward_hook(hook, *, prepend=False, with_kwargs=False, always_call=False) ``` -### [paddle.nn.Layer.register_forward_post_hook](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#register-forward-post-hook-hook) +### [paddle.nn.Module.register_forward_post_hook](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#register-forward-post-hook-hook) ```python -paddle.nn.Layer.register_forward_post_hook(hook) +paddle.nn.Module.register_forward_post_hook(hook) ``` PyTorch 参数更多，具体如下： diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_pre_hook.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_pre_hook.md index e295cffab71..21ae17aecd1 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_pre_hook.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.register_forward_pre_hook.md @@ -4,9 +4,9 @@ torch.nn.Module.register_forward_pre_hook(hook, *, prepend=False, with_kwargs=False) ``` -### [paddle.nn.Layer.register_forward_pre_hook](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#register-forward-pre-hook-hook) +### [paddle.nn.Module.register_forward_pre_hook](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer_cn.html#register-forward-pre-hook-hook) ```python -paddle.nn.Layer.register_forward_pre_hook(hook) +paddle.nn.Module.register_forward_pre_hook(hook) ``` PyTorch 参数更多，具体如下： diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.train.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.train.md deleted file mode 100644 index b18f4ffb8af..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.Module.train.md +++ /dev/null @@ -1,21 +0,0 @@ -## [ torch 参数更多 ]torch.nn.Module.train - -### [torch.nn.Module.train](https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.train) - -```python -torch.nn.Module.train(mode=True) -``` - -### [paddle.nn.Layer.train](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/Layer/train_cn.html#paddle/nn/Layer/train_cn#cn-api-paddle-nn-Layer-train) - -```python -paddle.nn.Layer.train() -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ---------- | ------------ | -- | -| `mode` | 无 | 设置训练模式，Paddle 无此参数，暂无转写方式。| diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.MultiheadAttention.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.MultiheadAttention.md deleted file mode 100644 index e8c6c407b17..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.MultiheadAttention.md +++ /dev/null @@ -1,30 +0,0 @@ -## [ torch 参数更多 ]torch.nn.MultiheadAttention -### [torch.nn.MultiheadAttention](https://pytorch.org/docs/stable/generated/torch.nn.MultiheadAttention.html#torch.nn.MultiheadAttention) -```python -torch.nn.MultiheadAttention(embed_dim, num_heads, dropout=0.0, bias=True, add_bias_kv=False, add_zero_attn=False, kdim=None, vdim=None, batch_first=False, device=None, dtype=None) -``` - -### [paddle.nn.MultiHeadAttention](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/MultiHeadAttention_cn.html) -```python -paddle.nn.MultiHeadAttention(embed_dim, num_heads, dropout=0.0, kdim=None, vdim=None, need_weights=False, weight_attr=None, bias_attr=None) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------------------------- | -| embed_dim | embed_dim | 输入输出的维度。 | -| num_heads | num_heads | 多头注意力机制的 Head 数量。 | -| dropout | dropout | 注意力目标的随机失活率。 | -| bias | bias_attr | 指定偏置参数属性的对象，Paddle 支持更多功能，同时支持 bool 用法。 | -| add_bias_kv | - | 在 dim=0 增加 bias，Paddle 无此参数，暂无转写方式。 | -| add_zero_attn | - | 在 dim=1 增加批量 0 值，Paddle 无此参数，暂无转写方式。 | -| kdim | kdim | 键值对中 key 的维度。 | -| vdim | vdim | 键值对中 value 的维度。 | -| batch_first | - | 表示输入数据的第 0 维是否代表 batch_size，Paddle 无此参数，暂无转写方式。 | -| device | - | Tensor 的设备，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| dtype | - | Tensor 的数据类型，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| - | need_weights | 表明是否返回注意力权重，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | weight_attr | 指定权重参数属性的对象，PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.kl_div.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.kl_div.md index 9f77c19419c..3e857c071bf 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.kl_div.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.kl_div.md @@ -23,6 +23,7 @@ paddle.nn.functional.kl_div(input, | PyTorch | PaddlePaddle | 备注 | | ------------ | ------------ | ------------------------------------------------------ | +| target | label | 与 input 对应的标签值，仅参数名不一致。 | | size_average | - | PyTorch 已弃用， Paddle 无此参数，需要转写。 | | reduce | - | PyTorch 已弃用， Paddle 无此参数，需要转写。 | | reduction | reduction | 表示对输出结果的计算方式。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.scaled_dot_product_attention.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.scaled_dot_product_attention.md deleted file mode 100644 index e577b19b552..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.scaled_dot_product_attention.md +++ /dev/null @@ -1,26 +0,0 @@ -## [ torch 参数更多 ]torch.nn.functional.scaled_dot_product_attention -### [torch.nn.functional.scaled_dot_product_attention](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html#torch-nn-functional-scaled-dot-product-attention) -```python -torch.nn.functional.scaled_dot_product_attention(query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, scale=None, enable_gqa=False) -``` - -### [paddle.nn.functional.scaled_dot_product_attention](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/functional/scaled_dot_product_attention_cn.html#scaled-dot-product-attention) -```python -paddle.nn.functional.scaled_dot_product_attention(query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, training=True, name=None) -``` - -两者功能基本一致，参数不一致，具体如下： - -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------- | -| query | query | 注意力模块中的查询张量。 | -| key | key | 注意力模块中的关键张量。 | -| value | value | 注意力模块中的值张量。 | -| attn_mask | attn_mask | 与添加到注意力分数的 `query`、 `key`、 `value` 类型相同的浮点掩码或者 bool 掩码, 默认值为 `None`。 | -| dropout_p | dropout_p | `dropout` 的比例, 默认值为 0.00 即不进行正则化。 | -| is_causal | is_causal | 是否启用因果关系, 默认值为 `False` 即不启用。 | -| scale | - | 在 softmax 之前应用的缩放因子。默认与 Paddle 行为一致。Paddle 无此参数，暂无转写方式。 | -| enable_gqa | - | 是否启用 GQA 优化支持。Paddle 无此参数，暂无转写方式。 | -| - | training | 是否处于训练阶段, 默认值为 `True` 即处于训练阶段。Pytorch 无此参数，默认行为等同与 `training=True`，Paddle 保持默认即可。| diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.smooth_l1_loss.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.smooth_l1_loss.md new file mode 100644 index 00000000000..241d17cd230 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.nn.functional.smooth_l1_loss.md @@ -0,0 +1,120 @@ +## [ torch 参数更多 ]torch.nn.functional.smooth_l1_loss + +### [torch.nn.functional.smooth_l1_loss](https://pytorch.org/docs/stable/generated/torch.nn.functional.smooth_l1_loss.html) + +```python +torch.nn.functional.smooth_l1_loss(input, + target, + size_average=None, + reduce=None, + reduction='mean', + beta=1.0) +``` + +### [paddle.nn.functional.smooth_l1_loss](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/functional/smooth_l1_loss_cn.html#smooth-l1-loss) + +```python +paddle.nn.functional.smooth_l1_loss(input, + label, + reduction='mean', + delta=1.0, + is_huber=True, + name=None) +``` + +Torch 中 Smooth L1 loss 的计算方式: + +$$ +\ell(x, y) = \left [l_1, ..., l_N\ \right ]^T +$$ + +其中: + +$$ +l_n = \begin{cases} +0.5 (x_n - y_n)^2 / beta, & \text{if } |x_n - y_n| < beta \\ +|x_n - y_n| - 0.5 * beta, & \text{otherwise } +\end{cases} +$$ + +而 Paddle 中 Smooth L1 loss 的计算方式: + +$$ +loss(x,y) = \left [ z_1, ..., z_N \right ]^T +$$ + +其中： + +$$ +z_i = \begin{cases} + 0.5(x_i - y_i)^2 & {if |x_i - y_i| < delta} \\ + delta * |x_i - y_i| - 0.5 * delta^2 & {otherwise} + \end{cases} +$$ + +当 $is\_huber$ 参数为 True 时，Paddle 为 huber 损失。 +当 $is\_huber$ 参数为 False 时，PyTorch 和 Paddle 计算过程一致，均为 huber 损失除以 $delta$ 值。 + + +两者功能一致，但 Paddle 的 `delta` 和 PyTorch 的 `beta` 参数在公式中用法不一致，具体如下： + +### 参数映射 + +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| input | input | 输入 Tensor。 | +| target | label | 输入 Tensor 对应的标签，仅参数名不一致。 | +| size_average | - | 已弃用，需要转写。 | +| reduce | - | 已弃用，需要转写。 | +| reduction | reduction | 表示应用于输出结果的规约方式，可选值有：'none', 'mean', 'sum'。 | +| beta | delta | SmoothL1Loss 损失的阈值参数。 | +| - | is_huber | 控制 huber_loss 与 smooth_l1_loss 的开关，Paddle 需设置为 False 。 | + + + + +### 转写示例 +#### size_average/reduce：对应到 reduction 为 sum +```python +# PyTorch 写法 +torch.nn.functional.smooth_l1_loss(input, label, size_average=False, reduce=True) +torch.nn.functional.smooth_l1_loss(input, label, size_average=False) + +# Paddle 写法 +paddle.nn.functional.smooth_l1_loss(input, label, reduction='sum') +``` + +#### size_average/reduce：对应到 reduction 为 mean +```python +# PyTorch 写法 +torch.nn.functional.smooth_l1_loss(input, label, size_average=True, reduce=True) +torch.nn.functional.smooth_l1_loss(input, label, reduce=True) +torch.nn.functional.smooth_l1_loss(input, label, size_average=True) +torch.nn.functional.smooth_l1_loss(input, label) + +# Paddle 写法 +paddle.nn.functional.smooth_l1_loss(input, label, reduction='mean') +``` + +#### size_average/reduce：对应到 reduction 为 none +```python +# PyTorch 写法 +torch.nn.functional.smooth_l1_loss(input, label, size_average=True, reduce=False) +torch.nn.functional.smooth_l1_loss(input, label, size_average=False, reduce=False) +torch.nn.functional.smooth_l1_loss(input, label, reduce=False) + +# Paddle 写法 +paddle.nn.functional.smooth_l1_loss(input, label, reduction='none') +``` + +#### beta +```python +# PyTorch 的 beta 参数转化为 delta 参数 +beta=0.8 + +# PyTorch 写法 +output = torch.nn.functional.smooth_l1_loss(input, label, beta=beta) + +# Paddle 写法 +output = paddle.nn.functional.smooth_l1_loss(input, label, delta=beta, is_huber=False) +``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.onnx.export.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.onnx.export.md index 8342effa295..24867feb91e 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.onnx.export.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.onnx.export.md @@ -31,7 +31,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | custom_opsets | - | 自定义 opset，Paddle 无此参数，暂无转写方式。 | | export_modules_as_functions | - | 是否导出模型为 functions，Paddle 无此参数，暂无转写方式。 | | - | input_spec | 描述存储模型 forward 方法的输入，PyTorch 无此参数，Paddle 保持默认即可。 | -| - | configs | 其他用于兼容的存储配置选项，PyTorch 无此参数，Paddle 保持默认即可。 | +| - | **configs | 其他用于兼容的存储配置选项，PyTorch 无此参数，Paddle 保持默认即可。 | ### 转写示例 #### f、model 参数转写 diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adadelta.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adadelta.md index 7a2cf03facc..cfd4b737ee7 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adadelta.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adadelta.md @@ -7,6 +7,8 @@ torch.optim.Adadelta(params, eps=1e-6, weight_decay=0, foreach=None, + *, + capturable=False, maximize=False, differentiable=False) ``` @@ -34,6 +36,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | eps | epsilon | 保持数值稳定性的短浮点类型值，仅参数名不一致。 | | weight_decay | weight_decay | 表示权重衰减系数，参数默认值不一致, PyTorch 默认为`0`， Paddle 默认为`None`，Paddle 需保持与 PyTorch 一致。 | | foreach | - | 是否使用优化器的 foreach 实现。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | +| capturable | - | 是否捕获优化器状态。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | maximize | - | 根据目标最大化参数，而不是最小化。Paddle 无此参数，暂无转写方式。 | | differentiable | - | 是否应通过训练中的优化器步骤进行自动微分。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | grad_clip | 梯度裁剪的策略。 PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adagrad.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adagrad.md index 27513b550c5..399c15460b0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adagrad.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adagrad.md @@ -8,8 +8,10 @@ torch.optim.Adagrad(params, initial_accumulator_value=0, eps=1e-10, foreach=None, + *, maximize=False, - differentiable=False) + differentiable=False, + fused=None) ``` ### [paddle.optimizer.Adagrad](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/Adagrad_cn.html) @@ -38,4 +40,5 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | foreach | - | 是否使用优化器的 foreach 实现。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | maximize | - | 根据目标最大化参数，而不是最小化。Paddle 无此参数，暂无转写方式。 | | differentiable | - | 是否应通过训练中的优化器步骤进行自动微分。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | +| fused | - | 表示是否使用 fused 的实现方式。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | grad_clip | 梯度裁剪的策略。 PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adam.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adam.md index a829f8d39cb..2577366da19 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adam.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adam.md @@ -7,11 +7,13 @@ torch.optim.Adam(params, eps=1e-08, weight_decay=0, amsgrad=False, + *, foreach=None, maximize=False, capturable=False, differentiable=False, - fused=None) + fused=None, + decoupled_weight_decay=False) ``` ### [paddle.optimizer.Adam](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/Adam_cn.html) @@ -47,6 +49,7 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | capturable | - | 在 CUDA 图中捕获此实例是否安全。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | differentiable | - | 是否应通过训练中的优化器步骤进行自动微分。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | fused | - | 是否使用融合实现（仅限 CUDA）。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | +| decoupled_weight_decay | - | 是否使用解耦权重衰减。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | grad_clip | 梯度裁剪的策略。 PyTorch 无此参数，Paddle 保持默认即可。 | | - | lazy_mode | 设为 True 时，仅更新当前具有梯度的元素。PyTorch 无此参数，Paddle 保持默认即可。 | | - | multi_precision | 是否在权重更新期间使用 multi-precision。PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.AdamW.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.AdamW.md index 3a5fc41c8ad..bab76898272 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.AdamW.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.AdamW.md @@ -7,6 +7,7 @@ torch.optim.AdamW(params, eps=1e-08, weight_decay=0.01, amsgrad=False, + *, maximize=False, foreach=None, capturable=False, diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adamax.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adamax.md index 9475e0dc578..751d24f4f7f 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adamax.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.Adamax.md @@ -7,8 +7,10 @@ torch.optim.Adamax(params, eps=1e-08, weight_decay=0, foreach=None, + *, maximize=False, - differentiable=False) + differentiable=False, + capturable=False) ``` ### [paddle.optimizer.Adamax](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/Adamax_cn.html) @@ -37,4 +39,5 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | foreach | - | 是否使用优化器的 foreach 实现。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | maximize | - | 根据目标最大化参数，而不是最小化。Paddle 无此参数，暂无转写方式。 | | differentiable | - | 是否应通过训练中的优化器步骤进行自动微分。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | +| capturable | - | 是否捕获优化器步骤的中间值。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | - | grad_clip | 梯度裁剪的策略。 PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.NAdam.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.NAdam.md index b3591664270..5dba076dd84 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.NAdam.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.NAdam.md @@ -8,6 +8,7 @@ torch.optim.NAdam(params, weight_decay=0, momentum_decay=0.004, decoupled_weight_decay=False, + *, foreach=None, maximize=False, capturable=False, diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.RAdam.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.RAdam.md index cba7fa9e1a9..f65b91d810a 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.RAdam.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.RAdam.md @@ -7,6 +7,7 @@ torch.optim.RAdam(params, eps=1e-08, weight_decay=0, decoupled_weight_decay=False, + *, foreach=None, maximize=False, capturable=False, diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.SGD.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.SGD.md index 1b68a875fdc..fd11598cb51 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.SGD.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.optim.SGD.md @@ -2,14 +2,16 @@ ### [torch.optim.SGD](https://pytorch.org/docs/stable/generated/torch.optim.SGD.html) ```python torch.optim.SGD(params, - lr, + lr=1e-3, momentum=0, dampening=0, weight_decay=0, nesterov=False, + *, maximize=False, foreach=None, - differentiable=False) + differentiable=False, + fused=None) ``` ### [paddle.optimizer.SGD](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/SGD_cn.html) @@ -36,4 +38,5 @@ PyTorch 相比 Paddle 支持更多其他参数，具体如下： | maximize | - | 根据目标最大化参数，而不是最小化。Paddle 无此参数，暂无转写方式。 | | foreach | - | 是否使用优化器的 foreach 实现。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | | differentiable | - | 是否应通过训练中的优化器步骤进行自动微分。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | +| fused | - | 是否使用融合内核。Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| | - | grad_clip | 梯度裁剪的策略。 PyTorch 无此参数，Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.scalar_tensor.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.scalar_tensor.md index 1c6f03210b3..b3bf3c087b5 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.scalar_tensor.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.scalar_tensor.md @@ -2,6 +2,7 @@ ### [torch.scalar_tensor](https://docs.pytorch.org/executorch/stable/extension-tensor.html#:~:text=can%20create%20a-,scalar%20tensor,-with%20scalar_tensor) ```python torch.scalar_tensor(s, + *, dtype=torch.float32, layout=torch.strided, device=None, diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.sparse.softmax.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.sparse.softmax.md index de890674e48..9fd739bccd0 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.sparse.softmax.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.sparse.softmax.md @@ -1,7 +1,7 @@ ## [ torch 参数更多 ]torch.sparse.softmax ### [torch.sparse.softmax](https://pytorch.org/docs/stable/generated/torch.sparse.softmax.html#torch.sparse.softmax) ```python -torch.sparse.softmax(input, dim, dtype=None) +torch.sparse.softmax(input, dim, *, dtype=None) ``` ### [paddle.sparse.nn.functional.softmax](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/sparse/nn/functional/softmax_cn.html) diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.special.erf.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.special.erf.md index b7b3e53251d..fc44a0909eb 100644 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.special.erf.md +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.special.erf.md @@ -2,6 +2,7 @@ ### [torch.special.erf](https://pytorch.org/docs/stable/special.html?highlight=torch+special+erf#torch.special.erf) ```python torch.special.erf(input, + *, out=None) ``` diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.unique.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.unique.md deleted file mode 100644 index 589cea49aba..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torch.unique.md +++ /dev/null @@ -1,33 +0,0 @@ -## [ torch 参数更多 ]torch.unique -### [torch.unique](https://pytorch.org/docs/stable/generated/torch.unique.html?highlight=unique#torch.unique) -```python -torch.unique(input, - sorted, - return_inverse, - return_counts, - dim=None) -``` - -### [paddle.unique](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/unique_cn.html#unique) -```python -paddle.unique(x, - return_index=False, - return_inverse=False, - return_counts=False, - axis=None, - dtype='int64', - name=None) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： -### 参数映射 - -| PyTorch | PaddlePaddle | 备注 | -| ------------- | ------------ | ------------------------------------------------------ | -| input | x | 表示输入 Tensor。 | -| sorted | - | 表示是否按升序返回排列，Paddle 无此参数。可直接删除，PyTorch 目前忽略了此参数。 | -| - | return_index | 表示是否返回独有元素在输入 Tensor 中的索引，PyTorch 无此参数。Paddle 保持默认即可。 | -| return_inverse| return_inverse| 表示是否返回输入 Tensor 的元素对应在独有元素中的索引。 | -| return_counts | return_counts| 表示是否返回每个独有元素在输入 Tensor 中的个数。 | -| dim | axis | 表示指定选取独有元素的轴。 | -| - | dtype | 表示返回值的类型，PyTorch 无此参数， Paddle 保持默认即可。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.Inception3.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.Inception3.md new file mode 100644 index 00000000000..ff5280fafee --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.Inception3.md @@ -0,0 +1,24 @@ +## [ torch 参数更多 ]torchvision.models.Inception3 +### [torchvision.models.Inception3](https://pytorch.org/vision/stable/models/generated/torchvision.models.inception_v3.html) +```python +torchvision.models.Inception3(num_classes: int = 1000, aux_logits: bool = True, transform_input: bool = False, inception_blocks: Optional[list[Callable[..., nn.Module]]] = None, init_weights: Optional[bool] = None, dropout: float = 0.5) +``` + +### [paddle.vision.models.InceptionV3](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/vision/models/InceptionV3_cn.html) +```python +paddle.vision.models.InceptionV3(num_classes: int = 1000, with_pool: bool = True) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| ---------------- | ------------ | -------------------------------------------------------------------- | +| num_classes | num_classes | 表示分类的类别数。参数完全一致。 | +| aux_logits | - | 是否使用辅助分类器，Paddle 无此参数，暂无转写方式。 | +| transform_input | - | 是否对输入进行预处理，Paddle 无此参数，暂无转写方式。 | +| inception_blocks | - | 用于构建网络的 Inception 模块，Paddle 无此参数，暂无转写方式。 | +| init_weights | - | 是否对权重进行初始化，Paddle 无此参数，暂无转写方式。 | +| dropout | - | Dropout 概率，Paddle 无此参数，暂无转写方式。 | +| - | with_pool | 是否在最后的全连接层前使用池化，Paddle 特有参数，使用默认值。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionA.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionA.md new file mode 100644 index 00000000000..b63d4e6371c --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionA.md @@ -0,0 +1,20 @@ +## [ torch 参数更多 ]torchvision.models.inception.InceptionA +### [torchvision.models.inception.InceptionA](https://pytorch.org/vision/stable/generated/torchvision.models.inception.InceptionA.html) +```python +torchvision.models.inception.InceptionA(in_channels: int, pool_features: int, conv_block: Optional[Callable[..., nn.Module]] = None) +``` + +### [paddle.vision.models.inceptionv3.InceptionA](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/vision/models/inceptionv3.py) +```python +paddle.vision.models.inceptionv3.InceptionA(num_channels: int, pool_features: int) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| :------------ | :----------- | :----------------------------------------------------------- | +| in_channels | num_channels | 输入通道数，仅参数名不一致。 | +| pool_features | pool_features| 混合层中池化层后的卷积输出通道数。 | +| conv_block | - | 用于自定义卷积模块，Paddle 无此参数，默认使用 standard conv，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionB.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionB.md new file mode 100644 index 00000000000..2947f8bb14b --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionB.md @@ -0,0 +1,19 @@ +## [ torch 参数更多 ]torchvision.models.inception.InceptionB +### [torchvision.models.inception.InceptionB](https://pytorch.org/vision/stable/generated/torchvision.models.inception.InceptionB.html) +```python +torchvision.models.inception.InceptionB(in_channels: int, conv_block: Optional[Callable[..., nn.Module]] = None) +``` + +### [paddle.vision.models.inceptionv3.InceptionB](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/vision/models/inceptionv3.py) +```python +paddle.vision.models.inceptionv3.InceptionB(num_channels: int) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| :------------ | :----------- | :----------------------------------------------------------- | +| in_channels | num_channels | 输入通道数，仅参数名不一致。 | +| conv_block | - | 用于自定义卷积模块，Paddle 无此参数，默认使用 standard conv，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionC.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionC.md new file mode 100644 index 00000000000..6193fbc7283 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionC.md @@ -0,0 +1,20 @@ +## [ torch 参数更多 ]torchvision.models.inception.InceptionC +### [torchvision.models.inception.InceptionC](https://pytorch.org/vision/stable/generated/torchvision.models.inception.InceptionC.html) +```python +torchvision.models.inception.InceptionC(in_channels: int, channels_7x7: int, conv_block: Optional[Callable[..., nn.Module]] = None) +``` + +### [paddle.vision.models.inceptionv3.InceptionC](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/vision/models/inceptionv3.py) +```python +paddle.vision.models.inceptionv3.InceptionC(num_channels: int, channels_7x7: int) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| :------------ | :----------- | :----------------------------------------------------------- | +| in_channels | num_channels | 输入通道数，仅参数名不一致。 | +| channels_7x7 | channels_7x7 | 7x7 卷积层的通道数。 | +| conv_block | - | 用于自定义卷积模块，Paddle 无此参数，默认使用 standard conv，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionD.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionD.md new file mode 100644 index 00000000000..e40f791953b --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionD.md @@ -0,0 +1,19 @@ +## [ torch 参数更多 ]torchvision.models.inception.InceptionD +### [torchvision.models.inception.InceptionD](https://pytorch.org/vision/stable/generated/torchvision.models.inception.InceptionD.html) +```python +torchvision.models.inception.InceptionD(in_channels: int, conv_block: Optional[Callable[..., nn.Module]] = None) +``` + +### [paddle.vision.models.inceptionv3.InceptionD](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/vision/models/inceptionv3.py) +```python +paddle.vision.models.inceptionv3.InceptionD(num_channels: int) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| :------------ | :----------- | :----------------------------------------------------------- | +| in_channels | num_channels | 输入通道数，仅参数名不一致。 | +| conv_block | - | 用于自定义卷积模块，Paddle 无此参数，默认使用 standard conv，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionE.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionE.md new file mode 100644 index 00000000000..646b014a5d6 --- /dev/null +++ b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/torchvision.models.inception.InceptionE.md @@ -0,0 +1,19 @@ +## [ torch 参数更多 ]torchvision.models.inception.InceptionE +### [torchvision.models.inception.InceptionE](https://pytorch.org/vision/stable/generated/torchvision.models.inception.InceptionE.html) +```python +torchvision.models.inception.InceptionE(in_channels: int, conv_block: Optional[Callable[..., nn.Module]] = None) +``` + +### [paddle.vision.models.inceptionv3.InceptionE](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/vision/models/inceptionv3.py) +```python +paddle.vision.models.inceptionv3.InceptionE(num_channels: int) +``` + +torchvision 相比 Paddle 支持更多其他参数，具体如下： + +### 参数映射 + +| torchvision | PaddlePaddle | 备注 | +| :------------ | :----------- | :----------------------------------------------------------- | +| in_channels | num_channels | 输入通道数，仅参数名不一致。 | +| conv_block | - | 用于自定义卷积模块，Paddle 无此参数，默认使用 standard conv，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.AddedToken.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.AddedToken.md deleted file mode 100644 index 2034a828dd6..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.AddedToken.md +++ /dev/null @@ -1,23 +0,0 @@ -## [ torch 参数更多 ]transformers.AddedToken -### [transformers.AddedToken](https://github.com/huggingface/transformers/blob/d625294d79341662784495551abdf45e6cb9372f/src/transformers/tokenization_utils_base.py#L84) -```python -transformers.AddedToken(content: str, single_word=False, lstrip=False, rstrip=False, special=False, normalized=None) -``` - -### [paddlenlp.transformers.AddedToken](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/transformers/tokenizer_utils_base.py#L48) -```python -paddlenlp.transformers.AddedToken(content: str = field(default_factory=str), single_word: bool = False， lstrip: bool = False, rstrip: bool = False, normalized: bool = True) -``` - -PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| transformers | PaddlePaddle | 备注 | -| ------------ | ------------ | ---------------------- | -| content | content | 待添加的 Token 内容。 | -| single_word | single_word | token 是否视为独立的词。 | -| lstrip | lstrip | 是否移除左侧空白符。 | -| rstrip | rstrip | 是否移除左侧空白符。 | -| special | - | 是否有特殊的处理方式，Paddle 暂不支持该参数。 | -| normalized | normalized | 是否进行规范化处理。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.GenerationConfig.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.GenerationConfig.md deleted file mode 100644 index 9e25ac758ff..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.GenerationConfig.md +++ /dev/null @@ -1,65 +0,0 @@ -## [ torch 参数更多 ]transformers.GenerationConfig -### [transformers.GenerationConfig](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/text_generation#transformers.GenerationConfig) -```python -transformers.GenerationConfig(*kwargs) -``` - -### [paddlenlp.generation.GenerationConfig](https://github.com/PaddlePaddle/PaddleNLP/blob/e336e78c338d2514ee6c937982ce5d8c960b85ff/paddlenlp/generation/configuration_utils.py#L62) -```python -paddlenlp.generation.GenerationConfig(*kwargs) -``` - -两者功能一致，但 PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| transformers | PaddlePaddle | 备注 | -| -------------------------------------| ------------------- | -------- | -| max_length | max_length | 最大生成长度。 | -| max_new_tokens | - | 最大生成长度(忽略 prompt)，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| min_length | min_length | 最小生成长度。 | -| min_new_tokens | - | 最小生成长度(忽略 prompt)，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| early_stopping | early_stopping | 早停是否开启。 | -| max_time | - | 最大允许计算运行时间，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| do_sample | do_sample | 是否进行采样。 | -| num_beams | num_beams | beams for beam search。 | -| num_beam_groups | num_beam_groups | beams 划分的组数。 | -| penalty_alpha | - | 惩罚参数，Paddle 无此参数，暂无转写方式。 | -| use_cache | use_cache | 是否开启 kv cache。 | -| temperature | temperature | 用于控制下个 token 生成的参数。 | -| top_k | top_k | top_k 算法的 k 值。 | -| top_p | top_p | top_p 算法的 p 值。 | -| typical_p | - | 局部典型度量的参数，Paddle 无此参数，暂无转写方式。 | -| epsilon_cutoff | - | 截断采样参数，Paddle 无此参数，暂无转写方式。 | -| eta_cutoff | - | 截断采样参数，Paddle 无此参数，暂无转写方式。 | -| diversity_penalty | - | 分散惩罚系数，Paddle 无此参数，暂无转写方式。 | -| repetition_penalty | repetition_penalty | 重复惩罚参数。 | -| encoder_repetition_penalty | - | 编码重复惩罚参数，Paddle 无此参数，暂无转写方式。 | -| length_penalty | length_penalty | 长度重复惩罚参数。 | -| no_repeat_ngram_size | - | ngram 在给定长度内不可重复，Paddle 无此参数，暂无转写方式。 | -| bad_words_ids | - | 不允许生成的 id 列表，Paddle 无此参数，暂无转写方式。 | -| force_words_ids | - | 必须生成的 id 列表，Paddle 无此参数，暂无转写方式。 | -| renormalize_logits | - | 对 logits 进行 renormalize 操作，Paddle 无此参数，暂无转写方式。 | -| constraints | - | 自定义约束列表，Paddle 无此参数，暂无转写方式。 | -| forced_bos_token_id | forced_bos_token_id | 强制生成的 bos token 的 id。 | -| forced_eos_token_id | forced_eos_token_id | 强制生成的 eos token 的 id。 | -| remove_invalid_values | - | 是否移除无效值，Paddle 无此参数，暂无转写方式。 | -| exponential_decay_length_penalty | - | 生成指定长度 tokens 后的惩罚参数，Paddle 无此参数，暂无转写方式。 | -| suppress_tokens | - | 生成期间被抑制的 tokens，Paddle 无此参数，暂无转写方式。 | -| begin_suppress_tokens | - | 开始生成时被抑制的 tokens，Paddle 无此参数，暂无转写方式。 | -| forced_decoder_ids | - | 指定 decoder 指定位置生成的 token，Paddle 无此参数，暂无转写方式。 | -| sequence_bias | - | 映射 sequence 到其偏执项的字典，Paddle 无此参数，暂无转写方式。 | -| guidance_scale | - | 控制 output 与 input 联系紧密程度的参数，Paddle 无此参数，暂无转写方式。 | -| low_memory | - | 使用更低显存占用的搜索算法，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| num_return_sequences | - | 为 batch 中每个序列独立计算返回序列的个数，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。| -| output_attentions | - | 是否返回注意力层的注意力张量，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| output_hidden_states | - | 是否返回注意力层的隐藏层张量，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| output_scores | - | 是否返回注意力层的得分张量，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| return_dict_in_generate | - | 是否返回 dict 而不是 tuple，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| pad_token_id | pad_token_id | padding token 的 id。 | -| bos_token_id | bos_token_id | beginning-of-sequence token 的 id。 | -| eos_token_id | eos_token_id | end-of-sequence token 的 id。 | -| encoder_no_repeat_ngram_size | - | encoder 中指定 size 内不能出现同一个 ngram，Paddle 无此参数，暂无转写方式。 | -| decoder_start_token_id | - | decoder 生成的第一个 token 的 id，Paddle 无此参数，暂无转写方式。 | -| num_assistant_tokens | - | 定义在每次迭代中由目标模型检查之前由辅助模型生成的推测令牌的数量，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| num_assistant_tokens_schedule | - | 定义推理时应更改最大辅助 tokens 的 schedule，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.PretrainedConfig.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.PretrainedConfig.md deleted file mode 100644 index 30e6f775c1a..00000000000 --- a/docs/guides/model_convert/convert_from_pytorch/api_difference/torch_more_args/transformers.PretrainedConfig.md +++ /dev/null @@ -1,66 +0,0 @@ -## [ torch 参数更多 ]transformers.PretrainedConfig -### [transformers.PretrainedConfig](https://hf-mirror.com/docs/transformers/v4.42.0/en/main_classes/configuration#transformers.PretrainedConfig) -```python -transformers.PretrainedConfig(*kwargs) -``` - -### [paddlenlp.transformers.PretrainedConfig](https://github.com/PaddlePaddle/PaddleNLP/blob/57000fa12ce67024238f0b56a6fde63c592c54ce/paddlenlp/transformers/configuration_utils.py#L317) -```python -paddlenlp.transformers.PretrainedConfig(*kwargs) -``` - -两者功能一致，但 PyTorch 相比 Paddle 支持更多其他参数，具体如下： - -### 参数映射 - -| transformers | PaddlePaddle | 备注 | -| -------------------------------------| ------------------------------ | -------- | -| name_or_path | name_or_path | 传递给 from_pretrained 的模型名称或模型所在路径。 | -| output_attentions | output_attentions | 是否返回注意力层的注意力张量。 | -| output_hidden_states | output_hidden_states | 是否返回注意力层的隐藏层张量。 | -| return_dict | return_dict | 是否返回 dict 而不是 tuple。 | -| is_encoder_decoder | is_encoder_decoder | 模型是否为 encoder-decoder 架构。 | -| is_decoder | min_length | 模型是否为 decoder only 架构。 | -| cross_attention_hidden_size | cross_attention_hidden_size | cross_attention 层隐藏层维数。 | -| add_cross_attention | add_cross_attention | 是否增加 cross_attention 层。 | -| tie_encoder_decoder | tie_encoder_decoder | encoder 与 decoder 的权重是否关联。| -| prune_heads | prune_heads | 修剪模型的 heads。 | -| chunk_size_feed_forward | chunk_size_feed_forward | 前馈层的 chunk size 。 | -| max_length | max_length | 最大生成长度。 | -| min_length | min_length | 最小生成长度。 | -| early_stopping | early_stopping | 早停是否开启。 | -| do_sample | do_sample | 是否进行采样。 | -| num_beams | num_beams | beams for beam search。 | -| num_beam_groups | num_beam_groups | beams 划分的组数。 | -| diversity_penalty | diversity_penalty | 分散惩罚系数。 | -| temperature | temperature | 用于控制下个 token 生成的参数。 | -| top_k | top_k | top_k 算法的 k 值。 | -| top_p | top_p | top_p 算法的 p 值。 | -| typical_p | - | 局部典型度量的参数，Paddle 无此参数，暂无转写方式。 | -| repetition_penalty | repetition_penalty | 重复惩罚参数。 | -| length_penalty | length_penalty | 长度重复惩罚参数。 | -| no_repeat_ngram_size | no_repeat_ngram_size | ngram 在给定长度内不可重复。 | -| encoder_no_repeat_ngram_size | encoder_no_repeat_ngram_size | encoder 中指定 size 内不能出现同一个 ngram。 | -| bad_words_ids | bad_words_ids | 不允许生成的 id 列表。 | -| num_return_sequences | num_return_sequences | 为 batch 中每个序列独立计算返回序列的个数。| -| output_scores | output_scores | 是否返回注意力层的得分张量，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| return_dict_in_generate | return_dict_in_generate | 是否返回 dict 而不是 tuple，Paddle 无此参数，一般对网络训练结果影响不大，可直接删除。 | -| forced_bos_token_id | forced_bos_token_id | 强制生成的 bos token 的 id。 | -| forced_eos_token_id | forced_eos_token_id | 强制生成的 eos token 的 id。 | -| remove_invalid_values | remove_invalid_values | 是否移除无效值。 | -| architectures | architectures | 可共用预训练参数的模型架构。 | -| finetuning_task | finetuning_task | 微调模型的任务名称。 | -| id2label | id2label | 索引到标签的映射。 | -| label2id | label2id | 标签到索引的映射。 | -| num_labels | num_labels | 模型最后一层使用的标签数。 | -| task_specific_params | task_specific_params | 为当前任务额外指定的参数。 | -| problem_type | problem_type | 用于解决的问题类型。 | -| tokenizer_class | tokenizer_class | tokenizer 的类别。 | -| prefix | prefix | 调用前增加的 prompt。 | -| pad_token_id | pad_token_id | padding token 的 id。 | -| bos_token_id | bos_token_id | beginning-of-sequence token 的 id。 | -| eos_token_id | eos_token_id | end-of-sequence token 的 id。 | -| decoder_start_token_id | decoder_start_token_id | decoder 生成的第一个 token 的 id。 | -| torchscript | - | 模型是否和 torchscript 一起使用，Paddle 无此参数，暂无转写方式。| -| tie_word_embeddings | tie_word_embeddings | input 和 output 的 word embedding 层参数是否绑定。 | -| torch_dtype | dtype | 模型参数的数据类型。 | diff --git a/docs/guides/model_convert/convert_from_pytorch/tools/get_api_difference_info.py b/docs/guides/model_convert/convert_from_pytorch/tools/get_api_difference_info.py index 47afda0b532..0ce897ff21d 100644 --- a/docs/guides/model_convert/convert_from_pytorch/tools/get_api_difference_info.py +++ b/docs/guides/model_convert/convert_from_pytorch/tools/get_api_difference_info.py @@ -80,6 +80,12 @@ def unescape_api(api): return api.replace(r"\_", "_") +def safe_remove_var_arg_start(arg: str) -> str: + if arg == "*": + return arg + return arg.lstrip("*") + + def split_args(args_str): """ 按逗号分割参数字符串，忽略括号内的逗号 @@ -428,8 +434,10 @@ def get_meta_from_diff_file( torch_arg, paddle_arg, note = args_table_content meta_data["args_mapping"].append( { - "src_arg": torch_arg, - "dst_arg": paddle_arg, + "src_arg": safe_remove_var_arg_start(torch_arg), + "dst_arg": safe_remove_var_arg_start( + paddle_arg + ), "note": note, } ) @@ -626,40 +634,78 @@ def get_table_header_by_prefix(prefix): def discover_all_metas(cfp_basedir): - # 获取 api_difference/ 下的 api 映射文档 - diff_3rd_basedir = os.path.join(cfp_basedir, "api_difference") - - diff_srcs = [("api_difference", "torch.", "paddle.")] - diff_srcs += [ - ("api_difference", "fairscale.", "paddle."), - ("api_difference", "flash_attn.", "paddle."), - ("api_difference", "transformers.", "paddlenlp."), - ("api_difference", "torchvision.", ""), - ] + """ + 递归扫描 api_difference 目录下所有的 .md 文件，无视目录结构， + 根据文件名自动推断所属库并解析。 + """ + search_root = os.path.join(cfp_basedir, "api_difference") + if not os.path.exists(search_root): + print(f"Error: Directory not found: {search_root}") + return [] - diff_files = [] - for diff_src, api_prefix, dst_prefix in diff_srcs: - basedir = os.path.join(cfp_basedir, diff_src) - files = discover_markdown_files(basedir, api_prefix) + metas = [] + print(f"Recursively scanning all .md files in: {search_root} ...") + + for root, dirs, files in os.walk(search_root): + for filename in files: + # 1. 基础过滤：只看 md 文件，且忽略 README + if not filename.endswith(".md") or filename.lower() == "readme.md": + continue + + # 2. 根据文件名推断 src_prefix 和 dst_prefix + # 逻辑：文件名通常是 "库名.模块.API.md" + src_prefix = "" + dst_prefix = "" + + if filename.startswith("torchvision."): + src_prefix = "torchvision." + dst_prefix = "" # torchvision 映射通常留空或视具体情况而定 + elif filename.startswith("transformers."): + src_prefix = "transformers." + dst_prefix = "paddleformers." + elif filename.startswith("fairscale."): + src_prefix = "fairscale." + dst_prefix = "paddle." + elif filename.startswith("flash_attn."): + src_prefix = "flash_attn." + dst_prefix = "paddle." + elif filename.startswith("torch."): + src_prefix = "torch." + dst_prefix = "paddle." + elif filename.startswith("os."): + src_prefix = "os." + dst_prefix = "paddle." + elif filename.startswith("setuptools."): + src_prefix = "setuptools." + dst_prefix = "paddle." + else: + # 如果文件名不符合任何已知前缀，打印警告但尝试按 torch 处理（或跳过） + # 这里选择跳过，防止解析非 API 文档报错 + print(f"Skipping unrecognized file prefix: {filename}") + continue - # 新增过滤逻辑：跳过包含"others"的文件路径 - filtered_files = [f for f in files if "others" not in f] - diff_files.append(((api_prefix, dst_prefix), filtered_files)) + filepath = os.path.join(root, filename) - metas = [] - for prefixs, files in diff_files: - s, d = prefixs - sh = get_table_header_by_prefix(s) - for f in files: - # 确保文件路径中不包含"others"才处理 - if "others" not in f: - metas.append( - get_meta_from_diff_file(f, s, d, src_argmap_title=sh) + # 3. 解析文件 + try: + src_header = get_table_header_by_prefix(src_prefix) + + # 调用原有的解析函数 + meta = get_meta_from_diff_file( + filepath, + src_prefix, + dst_prefix, + src_argmap_title=src_header, ) + metas.append(meta) + except Exception as e: + # 打印错误但不中断整个流程，方便排查具体坏文件 + print(f"Warning: Failed to parse {filepath}") + print(f" Error: {e}") metas.sort(key=lambda x: x["src_api"]) print( - f"extracted {len(metas)} mapping metas data (excluding 'others' files)." + f"Successfully extracted {len(metas)} mapping metas data (No filtering)." ) return metas diff --git a/docs/guides/model_convert/convert_from_pytorch/tools/validate_api_difference_consistency.py b/docs/guides/model_convert/convert_from_pytorch/tools/validate_api_difference_consistency.py index fdb6b778506..c5617719a6c 100644 --- a/docs/guides/model_convert/convert_from_pytorch/tools/validate_api_difference_consistency.py +++ b/docs/guides/model_convert/convert_from_pytorch/tools/validate_api_difference_consistency.py @@ -3,6 +3,36 @@ from utils import extract_no_need_convert_list, load_mapping_json +# Note(littleherozzzx): get_api_difference_info.py not support parse overloaded +# functions currently. Currently, we hard code the check of overloaded functions +# in this file. + +TENSOR_SPLIT_SIG_ARGS = [("sections",), ("indices",)] +SPLIT_SIG_ARGS = [("input", "sections"), ("input", "indices")] +REDUCE_ARGS = [("dim", "keepdim"), ("other",)] +OVERLOADED_APIS = { + "torch.Tensor.dsplit": {"src_args": TENSOR_SPLIT_SIG_ARGS}, + "torch.Tensor.hsplit": {"src_args": TENSOR_SPLIT_SIG_ARGS}, + "torch.dsplit": {"src_args": SPLIT_SIG_ARGS}, + "torch.hsplit": {"src_args": SPLIT_SIG_ARGS}, + "torch.vsplit": {"src_args": SPLIT_SIG_ARGS}, + "torch.Tensor.vsplit": {"src_args": TENSOR_SPLIT_SIG_ARGS}, + "torch.Tensor.max": {"src_args": REDUCE_ARGS}, + "torch.Tensor.min": {"src_args": REDUCE_ARGS}, +} + + +ALLOW_MISSING_DIFF_DOCS = [ + # Flags 类API，使用 Mock 实现，对 Paddle 行为无影响，归类为可删去。 + "torch.backends.cuda.matmul.allow_tf32", + "torch.backends.cudnn.allow_tf32", + "torch.backends.cudnn.benchmark", + "torch.backends.cudnn.deterministic", + "torch.backends.cudnn.enabled", + # 不支持转换的 API + "torch.Tensor.rename", +] + def validate_api_mappings(): sum = 0 @@ -11,6 +41,7 @@ def validate_api_mappings(): # 加载API差异数据 api_diff = load_mapping_json(current_dir / "api_difference_info.json") + api_diff_map: dict[str:dict] = {} # 加载API映射数据 api_map = load_mapping_json(current_dir / "api_mapping.json") @@ -19,6 +50,9 @@ def validate_api_mappings(): api_map = api_map | attr_map + for api in ALLOW_MISSING_DIFF_DOCS: + api_map.pop(api, None) + no_need_list = extract_no_need_convert_list( str(current_dir) + "/global_var.py" ) @@ -26,64 +60,110 @@ def validate_api_mappings(): # 准备错误报告文件 error_file = current_dir / "validate_api_difference_consistency_error.txt" - # 任务1: 检查api_mapping中存在Matcher的条目是否在差异文档中有对应 with open(error_file, "w", encoding="utf-8") as err_file: + # 任务0: 检查 api_diff 文档中 src_api 是唯一的，并构造映射字典 + for entry in api_diff: + src_api = entry["src_api"] + if src_api in api_diff_map: + err_file.write( + f"ERROR: api_difference_info entry '{src_api}' is not unique\n" + ) + sum += 1 + else: + api_diff_map[src_api] = entry + + # 任务1: 检查api_mapping中存在Matcher的条目是否在差异文档中有对应 for api_key, api_value in api_map.items(): if ( api_value.get("Matcher") is not None and api_key not in no_need_list ): - if not any(entry["src_api"] == api_key for entry in api_diff): + if api_key not in api_diff_map: err_file.write( f"WARNING: api_mapping entry '{api_key}' not found in api_difference_info.json\n" ) sum += 1 # 任务2: 检查差异文档中的dst_api是否与api_mapping中paddle_api一致 - for entry in api_diff: - if "dst_api" not in entry: + for diff_entry in api_diff: + # 如果文档里没写目标 API（可能是组合实现），跳过 + if "dst_api" not in diff_entry: continue - found = False - for api_key, api_value in api_map.items(): - if api_value.get("paddle_api") == entry["dst_api"]: - found = True - break + torch_api = diff_entry["src_api"] + doc_target = diff_entry["dst_api"] # 文档里的目标 + + # 1. 检查代码规则库里有没有这个 torch_api + # 注意：这里要查 api_map (规则)，而不是 api_diff_map (文档自己) + if torch_api not in api_map: + # 这种情况（文档有，规则无）通常不在这里处理，或者仅作为 Info + continue + + rule_entry = api_map[torch_api] + + # 2. 获取代码规则里的目标 API + # 注意：api_mapping.json 里的键名是 'paddle_api' + code_target = rule_entry.get("paddle_api") - if not found: + # 3. 如果代码里有明确的目标 API，检查是否一致 + if code_target and code_target != doc_target: err_file.write( - f"WARNING: api_difference_info entry '{entry['src_api']}' has dst_api '{entry['dst_api']}' not found in api_mapping.json\n" + f"ERROR: Mapping Conflict for '{torch_api}':\n" + f" Document says -> {doc_target}\n" + f" Code rule says -> {code_target}\n" ) sum += 1 + # for entry in api_diff: + # if "dst_api" not in entry: + # continue + + # found = False + # for api_key, api_value in api_map.items(): + # if api_value.get("paddle_api") == entry["dst_api"]: + # found = True + # break + + # if not found: + # err_file.write( + # f"WARNING: api_difference_info entry '{entry['src_api']}' has dst_api '{entry['dst_api']}' not found in api_mapping.json\n" + # ) + # sum += 1 + # 任务3: 检查api_mapping中的kwargs_change是否在api_difference_info中 for api_key, api_value in api_map.items(): if "kwargs_change" not in api_value or api_key in no_need_list: continue + if api_key in OVERLOADED_APIS: # 重载暂不支持解析 diff 文档 + continue + entry = api_diff_map.get(api_key) + if entry is None: + continue + if entry["mapping_type"] == "组合替代实现": + continue # 组合替代实现不需要参数映射 for src_arg, dst_arg in api_value["kwargs_change"].items(): found = False - for entry in api_diff: - if "args_mapping" not in entry: - continue - for mapping in entry["args_mapping"]: - if mapping["dst_arg"] == "-": - mapping["dst_arg"] = "" - if "," in mapping["dst_arg"]: - mapping["dst_arg"] = ( - mapping["dst_arg"] - .replace(" ", "") - .replace("\t", "") - .split(",") - ) - if ( - mapping["src_arg"] == src_arg - and mapping["dst_arg"] == dst_arg - ): - found = True - break - if found: + if "args_mapping" not in entry: + continue + for mapping in entry["args_mapping"]: + if mapping["dst_arg"] == "-": + mapping["dst_arg"] = "" + if "," in mapping["dst_arg"]: + mapping["dst_arg"] = ( + mapping["dst_arg"] + .replace(" ", "") + .replace("\t", "") + .split(",") + ) + if ( + mapping["src_arg"] == src_arg + and mapping["dst_arg"] == dst_arg + ): + found = True break + if found: + break if not found: err_file.write( @@ -91,20 +171,76 @@ def validate_api_mappings(): ) sum += 1 + # ----------------------------------------------------------------- + # 任务4: 检查api_mapping中的args_list是否都在api_difference_info的src_signature中 + # (支持 OVERLOADED_APIS) + # ----------------------------------------------------------------- + # ----------------------------------------------------------------- # 任务4: 检查api_mapping中的args_list是否都在api_difference_info的src_signature中 + # (支持 OVERLOADED_APIS，且双方统一清洗参数名) + # ----------------------------------------------------------------- for api_key, api_value in api_map.items(): + # 跳过无需检查的情况 if "args_list" not in api_value or api_key in no_need_list: continue - # 获取api_difference_info中对应的entry - entry = next((e for e in api_diff if e["src_api"] == api_key), None) + args_list_in_mapping = api_value["args_list"] + + # 定义清洗函数：去掉参数名前的 * 或 **，保留纯参数名 + def normalize_param_name(name): + if name == "*": + return "*" + return name.lstrip("*") + + # ------------------------------------------- + # 分支 A：硬编码的重载 API 检查 + # ------------------------------------------- + if api_key in OVERLOADED_APIS: + valid_signatures_groups = OVERLOADED_APIS[api_key].get( + "src_args", [] + ) + + # 1. 生成全集 (Reference): 清洗定义中的参数 + allowed_args_superset = set() + for group in valid_signatures_groups: + current_group = ( + (group,) if isinstance(group, str) else group + ) + cleaned_group = { + normalize_param_name(arg) for arg in current_group + } + allowed_args_superset.update(cleaned_group) + + # 2. 准备检查对象 (Target): 清洗 Mapping 中的参数 【修复点】 + mapping_args_set = { + normalize_param_name(arg) for arg in args_list_in_mapping + } + + # 3. 集合求差 + unknown_args = mapping_args_set - allowed_args_superset + + if unknown_args: + err_file.write( + f"ERROR: Args list {args_list_in_mapping} in api_mapping for OVERLOADED API '{api_key}' " + f"contains arguments (normalized: {unknown_args}) which are NOT found in any signature defined in OVERLOADED_APIS.\n" + ) + sum += 1 + continue + + # ------------------------------------------- + # 分支 B：基于 JSON 文档的普通 API 检查 + # ------------------------------------------- + entry = api_diff_map.get(api_key) if entry is None: err_file.write( - f"ERROR: api_mapping for '{api_key}' not found in api_difference_info.json, so cannot check args_list\n" + f"ERROR: api_mapping for '{api_key}' not found in api_difference_info.json\n" ) sum += 1 continue + if entry.get("mapping_type") == "组合替代实现": + continue + if "src_signature" not in entry or not entry["src_signature"]: err_file.write( f"ERROR: api_difference_info for '{api_key}' has no src_signature\n" @@ -112,7 +248,7 @@ def validate_api_mappings(): sum += 1 continue - # 提取第一个src_signature的参数名列表 + # 1. 提取并清洗文档中的签名 (Reference) src_signature = entry["src_signature"][0] if "args" not in src_signature: err_file.write( @@ -120,16 +256,23 @@ def validate_api_mappings(): ) sum += 1 continue - src_args = [arg["arg_name"] for arg in src_signature["args"]] - # 检查api_mapping中的args_list是否都在src_args中 - for arg in api_value["args_list"]: - if arg not in src_args: + src_args = [ + normalize_param_name(arg["arg_name"]) + for arg in src_signature["args"] + ] + + # 2. 逐个检查 Mapping 中的参数 (Target) + for arg in args_list_in_mapping: + # 【修复点】：先清洗，再比较 + normalized_arg = normalize_param_name(arg) + + if normalized_arg not in src_args: err_file.write( - f"ERROR: Parameter '{arg}' in api_mapping for '{api_key}' not found in api_difference_info's src_signature\n" + f"ERROR: Parameter '{arg}' (normalized: '{normalized_arg}') in api_mapping for '{api_key}' " + f"not found in api_difference_info's src_signature: {src_args}\n" ) sum += 1 - print( f"{sum} api error found in api_mapping.json and api_difference_info.json" )