Skip to content

Commit 8a41737

Browse files
SamitHuangSamitHuang
committed
update training time in readme
1 parent 1aff357 commit 8a41737

File tree

4 files changed

+19
-18
lines changed

4 files changed

+19
-18
lines changed

‎configs/det/dbnet/README.md‎

Lines changed: 5 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -37,15 +37,16 @@ The overall architecture of DBNet is presented in _Figure 1._ It consists of mul
3737
### ICDAR2015
3838
<div align="center">
3939

40-
| **Model** | **Context** | **Backbone** | **Pretrained** | **Recall** | **Precision** | **F-score** | **Recipe** | **Download** |
41-
|---------------|--------------|----------------|------------|---------------|-------------|-----------------------------|----------------------------------------------------------------------------------------------| ----------|
42-
| DBNet (ours) | D910x1-MS1.9-G | ResNet-50 | ImageNet | 81.70% | 85.84% | 83.72% | [yaml](db_r50_icdar15.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/dbnet/dbnet_resnet50-db1df47a.ckpt) |
43-
| DBNet (PaddleOCR) | - | ResNet50_vd | SynthText | 78.72% | 86.41% | 82.38% |
40+
| **Model** | **Context** | **Backbone** | **Pretrained** | **Recall** | **Precision** | **F-score** | **Train T. (s/epoch)** | **Recipe** | **Download** |
41+
|---------------|--------------|----------------|------------|---------------|-------------|-----------------------------|-------------------------------------------|---------------------------------------------------| ----------|
42+
| DBNet (ours) | D910x1-MS1.9-G | ResNet-50 | ImageNet | 81.70% | 85.84% | 83.72% | 35 | [yaml](db_r50_icdar15.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/dbnet/dbnet_resnet50-db1df47a.ckpt) |
43+
| DBNet (PaddleOCR) | - | ResNet50_vd | SynthText | 78.72% | 86.41% | 82.38% | - | - | -|
4444

4545
</div>
4646

4747
#### Notes
4848
- Context: Training context denoted as {device}x{pieces}-{MS version}{MS mode}, where mindspore mode can be G - graph mode or F - pynative mode with ms function. For example, D910x8-G is for training on 8 pieces of Ascend 910 NPU using graph mode.
49+
- Note that the training time of DBNet is highly affected by data processing and varies on different machines.
4950

5051

5152

‎configs/det/dbnet/README_CN.md‎

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -28,10 +28,10 @@ DBNet的整体架构图如图1所示，包含以下阶段:
2828
### ICDAR2015
2929
<div align="center">
3030

31-
| **模型** | **环境配置** | **骨干网络** | **预训练数据集** | **Recall** | **Precision** | **F-score** | **配置文件** | **模型权重下载** |
32-
|------------------|------------|-------------|----------------|------------|---------------|-------------|-------------------------------|----------------------------------------------------------------------------------------------|
33-
| DBNet (ours) | D910x1-MS1.9-G | ResNet-50 | ImageNet | 81.70% | 85.84% | 83.72% | [yaml](db_r50_icdar15.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/dbnet/dbnet_resnet50-db1df47a.ckpt) |
34-
| DBNet (PaddleOCR)| - | ResNet50_vd | SynthText | 78.72% | 86.41% | 82.38% |
31+
| **模型** | **环境配置** | **骨干网络** | **预训练数据集** | **Recall** | **Precision** | **F-score** | **训练时间(s/epoch)** | **配置文件** | **模型权重下载** |
32+
|------------------|------------|-------------|----------------|------------|---------------|-------------|-------------------------------|-----------------------------|-----------------------------------------------------------------|
33+
| DBNet (ours) | D910x1-MS1.9-G | ResNet-50 | ImageNet | 81.70% | 85.84% | 83.72% | 35 | [yaml](db_r50_icdar15.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/dbnet/dbnet_resnet50-db1df47a.ckpt) |
34+
| DBNet (PaddleOCR)| - | ResNet50_vd | SynthText | 78.72% | 86.41% | 82.38% | - | - | -|
3535

3636
</div>
3737

‎configs/rec/crnn/README.md‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -37,11 +37,11 @@ According to our experiments, the evaluation results on public benchmark dataset
3737

3838
<div align="center">
3939

40-
| **Model** | **Context** | **Backbone** | **Avg Accuracy** | **Recipe** | **Download** |
41-
|-----------|--------------|------------------|------------|--------------| ------ |
42-
| CRNN (ours) | D910x8-MS1.8-G | VGG7 | 82.03% | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_vgg7.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_vgg7-ea7e996c.ckpt) |
43-
| CRNN (ours) | D910x8-MS1.8-G | ResNet34_vd | 84.45% | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_resnet34.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_resnet34-83f37f07.ckpt) |
44-
| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | [yaml](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/configs/rec/rec_r34_vd_none_bilstm_ctc.yml) | [weights](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar) |
40+
| **Model** | **Context** | **Backbone** | **Avg Accuracy** | **Train T. (s/epoch)** | **Recipe** | **Download** |
41+
|-----------|--------------|------------------|------------|--------------| ------ |------ |
42+
| CRNN (ours) | D910x8-MS1.8-G | VGG7 | 82.03% | 2445 | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_vgg7.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_vgg7-ea7e996c.ckpt) |
43+
| CRNN (ours) | D910x8-MS1.8-G | ResNet34_vd | 84.45% | 2118 | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_resnet34.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_resnet34-83f37f07.ckpt) |
44+
| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | -| -| - |
4545

4646
</div>
4747

‎configs/rec/crnn/README_CN.md‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -37,11 +37,11 @@ Table Format:
3737

3838
<div align="center">
3939

40-
| **模型** | **环境配置** |**骨干网络** | **平均准确率** | **配置文件** | **模型权重下载** |
41-
|-----------|--------------|------------------|------------|--------------| ------ |
42-
| CRNN (ours) | D910x8-MS1.8-G | VGG7 | 82.03% | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_vgg7.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_vgg7-ea7e996c.ckpt) |
43-
| CRNN (ours) | D910x8-MS1.8-G | ResNet34_vd | 84.45% | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_resnet34.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_resnet34-83f37f07.ckpt) |
44-
| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | [yaml](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/configs/rec/rec_r34_vd_none_bilstm_ctc.yml) | [weights](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar) |
40+
| **模型** | **环境配置** |**骨干网络** | **平均准确率** | 训练时间(s/epoch) | **配置文件** | **模型权重下载** |
41+
|-----------|--------------|------------------|------------|--------------| ------ | ------|
42+
| CRNN (ours) | D910x8-MS1.8-G | VGG7 | 82.03% | 2445 | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_vgg7.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_vgg7-ea7e996c.ckpt) |
43+
| CRNN (ours) | D910x8-MS1.8-G | ResNet34_vd | 84.45% | 2118 | [yaml](https://github.com/mindspore-lab/mindocr/blob/main/configs/rec/crnn/crnn_resnet34.yaml) | [weights](https://download.mindspore.cn/toolkits/mindocr/crnn/crnn_resnet34-83f37f07.ckpt) |
44+
| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | - | - |
4545

4646
</div>
4747

0 commit comments

Comments
 (0)