Merge pull request #150 from SamitHuang/version

udpate model readme with training time

Author: SamitHuang

configs/det/dbnet/README.md

@@ -37,15 +37,16 @@ The overall architecture of DBNet is presented in _Figure 1._ It consists of mul
 ### ICDAR2015
 <div align="center">
 
-| **Model** |  **Context** | **Backbone** | **Pretrained** | **Recall** | **Precision** | **F-score** | **Recipe**  | **Download**  |
-|---------------|--------------|----------------|------------|---------------|-------------|-----------------------------|----------------------------------------------------------------------------------------------| ----------|
-| 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) |
-| DBNet (PaddleOCR) | - | ResNet50_vd  | SynthText      | 78.72%     | 86.41%        | 82.38%      |
+| **Model** |  **Context** | **Backbone** | **Pretrained** | **Recall** | **Precision** | **F-score** | **Train T. (s/epoch)** | **Recipe**  | **Download**  |
+|---------------|--------------|----------------|------------|---------------|-------------|-----------------------------|-------------------------------------------|---------------------------------------------------| ----------|
+| 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) |
+| DBNet (PaddleOCR) | - | ResNet50_vd  | SynthText      | 78.72%     | 86.41%        | 82.38%      | - | - | -|
 
 </div>
 
 #### Notes
 - 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.
+- Note that the training time of DBNet is highly affected by data processing and varies on different machines. 
 
 
 

configs/det/dbnet/README_CN.md

@@ -28,10 +28,10 @@ DBNet的整体架构图如图1所示，包含以下阶段:
 ### ICDAR2015
 <div align="center">
 
-| **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**                                                                               |
-|------------------|------------|-------------|----------------|------------|---------------|-------------|-------------------------------|----------------------------------------------------------------------------------------------|
-| 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) |
-| DBNet (PaddleOCR)|  - | ResNet50_vd  | SynthText      | 78.72%     | 86.41%        | 82.38%      |
+| **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**                                                                               |
+|------------------|------------|-------------|----------------|------------|---------------|-------------|-------------------------------|-----------------------------|-----------------------------------------------------------------|
+| 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) |
+| DBNet (PaddleOCR)|  - | ResNet50_vd  | SynthText      | 78.72%     | 86.41%        | 82.38%      |  - | - | -|
 
 </div>
 

configs/rec/crnn/README.md

@@ -37,11 +37,11 @@ According to our experiments, the evaluation results on public benchmark dataset
 
 <div align="center">
 
-| **Model** | **Context** | **Backbone** | **Avg Accuracy** | **Recipe** | **Download** | 
-|-----------|--------------|------------------|------------|--------------| ------ |
-| 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)     |
-| 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) |
-| 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) |
+| **Model** | **Context** | **Backbone** | **Avg Accuracy** | **Train T. (s/epoch)** | **Recipe** | **Download** | 
+|-----------|--------------|------------------|------------|--------------| ------ |------ |
+| 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)     |
+| 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) |
+| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | -| -| - |
 
 </div>
 
@@ -50,7 +50,7 @@ According to our experiments, the evaluation results on public benchmark dataset
 - Both VGG and ResNet models are trained from scratch without any pre-training.
 - The above models are trained with MJSynth (MJ) and SynthText (ST) datasets. For more data details, please refer to [Data Preparation](#dataset-preparation)
 - Evaluations are tested individually on each benchmark dataset, and Avg Accuracy is the average of accuracies across all sub-datasets.
-- PaddleOCR version of CRNN, we directly use the trained model provided on their [github](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/doc/doc_en/algorithm_rec_crnn_en.md).
+- For the PaddleOCR version of CRNN, the performance is reported on the trained model provided on their [github](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/doc/doc_en/algorithm_rec_crnn_en.md).
 
 
 ## Quick Start

configs/rec/crnn/README_CN.md

@@ -37,11 +37,11 @@ Table Format:
 
 <div align="center">
 
-| **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** | 
-|-----------|--------------|------------------|------------|--------------| ------ |
-| 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)     |
-| 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) |
-| 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) |
+| **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** | 
+|-----------|--------------|------------------|------------|--------------| ------ | ------|
+| 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)     |
+| 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) |
+| CRNN (PaddleOCR) | - | ResNet34_vd | 83.99% | - | - |
 
 </div>