Skip to content

GitLab

"vscode:/vscode.git/clone" did not exist on "d9316bf8bc742dfa7635cf0cb2d5fea0cd1b0d00"
Unverified Commit 90b968d5 authored by xiaoting's avatar xiaoting Committed by GitHub
Browse files

Merge pull request #1317 from xmy0916/dygraph 

fix doc algorithm_overview ch&en
parents e83bd2d8 77376859
Hide whitespace changes
Inline Side-by-side
Showing with 191 additions and 129 deletions
doc/doc_ch/algorithm_overview.md
View file @ 90b968d5
...@@ -17,17 +17,17 @@ PaddleOCR开源的文本检测算法列表： ...@@ -17,17 +17,17 @@ PaddleOCR开源的文本检测算法列表：
|模型|骨干网络|precision|recall|Hmean|下载链接| |模型|骨干网络|precision|recall|Hmean|下载链接|
|-|-|-|-|-|-| |-|-|-|-|-|-|
|EAST|ResNet50_vd|88.18%|85.51%|86.82%|[下载链接](https://paddleocr.bj.bcebos.com/det_r50_vd_east.tar)| |EAST|ResNet50_vd||||[敬请期待]()|
|EAST|MobileNetV3|81.67%|79.83%|80.74%|[下载链接](https://paddleocr.bj.bcebos.com/det_mv3_east.tar)| |EAST|MobileNetV3||||[敬请期待]()|
|DB|ResNet50_vd|83.79%|80.65%|82.19%|[下载链接](https://paddleocr.bj.bcebos.com/det_r50_vd_db.tar)| |DB|ResNet50_vd||||[敬请期待]()|
|DB|MobileNetV3|75.92%|73.18%|74.53%|[下载链接](https://paddleocr.bj.bcebos.com/det_mv3_db.tar)| |DB|MobileNetV3||||[敬请期待]()|
|SAST|ResNet50_vd|92.18%|82.96%|87.33%|[下载链接](https://paddleocr.bj.bcebos.com/SAST/sast_r50_vd_icdar2015.tar)| |SAST|ResNet50_vd||||[敬请期待]()|
在Total-text文本检测公开数据集上，算法效果如下： 在Total-text文本检测公开数据集上，算法效果如下：
|模型|骨干网络|precision|recall|Hmean|下载链接| |模型|骨干网络|precision|recall|Hmean|下载链接|
|-|-|-|-|-|-| |-|-|-|-|-|-|
|SAST|ResNet50_vd|88.74%|79.80%|84.03%|[下载链接](https://paddleocr.bj.bcebos.com/SAST/sast_r50_vd_total_text.tar)| |SAST|ResNet50_vd||||[敬请期待]()|
**说明：** SAST模型训练额外加入了icdar2013、icdar2017、COCO-Text、ArT等公开数据集进行调优。PaddleOCR用到的经过整理格式的英文公开数据集下载：[百度云地址](https://pan.baidu.com/s/12cPnZcVuV1zn5DOd4mqjVw) (提取码: 2bpi) **说明：** SAST模型训练额外加入了icdar2013、icdar2017、COCO-Text、ArT等公开数据集进行调优。PaddleOCR用到的经过整理格式的英文公开数据集下载：[百度云地址](https://pan.baidu.com/s/12cPnZcVuV1zn5DOd4mqjVw) (提取码: 2bpi)
...@@ -48,17 +48,12 @@ PaddleOCR开源的文本识别算法列表： ...@@ -48,17 +48,12 @@ PaddleOCR开源的文本识别算法列表：
|模型|骨干网络|Avg Accuracy|模型存储命名|下载链接| |模型|骨干网络|Avg Accuracy|模型存储命名|下载链接|
|-|-|-|-|-| |-|-|-|-|-|
|Rosetta|Resnet34_vd|80.24%|rec_r34_vd_none_none_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_r34_vd_none_none_ctc.tar)| |Rosetta|Resnet34_vd||rec_r34_vd_none_none_ctc|[敬请期待]()|
|Rosetta|MobileNetV3|78.16%|rec_mv3_none_none_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_mv3_none_none_ctc.tar)| |Rosetta|MobileNetV3||rec_mv3_none_none_ctc|[敬请期待]()|
|CRNN|Resnet34_vd|82.20%|rec_r34_vd_none_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_r34_vd_none_bilstm_ctc.tar)| |CRNN|Resnet34_vd||rec_r34_vd_none_bilstm_ctc|[敬请期待]()|
|CRNN|MobileNetV3|79.37%|rec_mv3_none_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_mv3_none_bilstm_ctc.tar)| |CRNN|MobileNetV3||rec_mv3_none_bilstm_ctc|[敬请期待]()|
|STAR-Net|Resnet34_vd|83.93%|rec_r34_vd_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_r34_vd_tps_bilstm_ctc.tar)| |STAR-Net|Resnet34_vd||rec_r34_vd_tps_bilstm_ctc|[敬请期待]()|
|STAR-Net|MobileNetV3|81.56%|rec_mv3_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/rec_mv3_tps_bilstm_ctc.tar)| |STAR-Net|MobileNetV3||rec_mv3_tps_bilstm_ctc|[敬请期待]()|
|RARE|Resnet34_vd|84.90%|rec_r34_vd_tps_bilstm_attn|[下载链接](https://paddleocr.bj.bcebos.com/rec_r34_vd_tps_bilstm_attn.tar)|
|RARE|MobileNetV3|83.32%|rec_mv3_tps_bilstm_attn|[下载链接](https://paddleocr.bj.bcebos.com/rec_mv3_tps_bilstm_attn.tar)|
|SRN|Resnet50_vd_fpn|88.33%|rec_r50fpn_vd_none_srn|[下载链接](https://paddleocr.bj.bcebos.com/SRN/rec_r50fpn_vd_none_srn.tar)|
**说明：** SRN模型使用了数据扰动方法对上述提到对两个训练集进行增广，增广后的数据可以在[百度网盘](https://pan.baidu.com/s/1-HSZ-ZVdqBF2HaBZ5pRAKA)上下载，提取码: y3ry。
原始论文使用两阶段训练平均精度为89.74%，PaddleOCR中使用one-stage训练，平均精度为88.33%。两种预训练权重均在[下载链接](https://paddleocr.bj.bcebos.com/SRN/rec_r50fpn_vd_none_srn.tar)中。
PaddleOCR文本识别算法的训练和使用请参考文档教程中[模型训练/评估中的文本识别部分](./recognition.md)。 PaddleOCR文本识别算法的训练和使用请参考文档教程中[模型训练/评估中的文本识别部分](./recognition.md)。
doc/doc_ch/recognition.md
View file @ 90b968d5
...@@ -142,9 +142,8 @@ word_dict.txt 每行有一个单字，将字符与数字索引映射在一起， ...@@ -142,9 +142,8 @@ word_dict.txt 每行有一个单字，将字符与数字索引映射在一起，
<a name="支持空格"></a> <a name="支持空格"></a>
- 添加空格类别 - 添加空格类别
如果希望支持识别"空格"类别, 请将yml文件中的 `use_space_char` 字段设置为 `true`。 如果希望支持识别"空格"类别, 请将yml文件中的 `use_space_char` 字段设置为 `True`。
**注意：`use_space_char` 仅在 `character_type=ch` 时生效**
<a name="启动训练"></a> <a name="启动训练"></a>
### 启动训练 ### 启动训练
...@@ -167,10 +166,9 @@ tar -xf rec_mv3_none_bilstm_ctc.tar && rm -rf rec_mv3_none_bilstm_ctc.tar ...@@ -167,10 +166,9 @@ tar -xf rec_mv3_none_bilstm_ctc.tar && rm -rf rec_mv3_none_bilstm_ctc.tar
*如果您安装的是cpu版本，请将配置文件中的 `use_gpu` 字段修改为false* *如果您安装的是cpu版本，请将配置文件中的 `use_gpu` 字段修改为false*
``` ```
# GPU训练 支持单卡，多卡训练，通过CUDA_VISIBLE_DEVICES指定卡号 # GPU训练 支持单卡，多卡训练，通过--gpus参数指定卡号
export CUDA_VISIBLE_DEVICES=0,1,2,3
# 训练icdar15英文数据 并将训练日志保存为 tain_rec.log # 训练icdar15英文数据 并将训练日志保存为 tain_rec.log
python3 tools/train.py -c configs/rec/rec_icdar15_train.yml 2>&1 | tee train_rec.log python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/rec/rec_icdar15_train.yml
``` ```
<a name="数据增强"></a> <a name="数据增强"></a>
- 数据增强 - 数据增强
...@@ -195,8 +193,8 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t ...@@ -195,8 +193,8 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t
| 配置文件 | 算法名称 | backbone | trans | seq | pred | | 配置文件 | 算法名称 | backbone | trans | seq | pred |
| :--------: | :-------: | :-------: | :-------: | :-----: | :-----: | | :--------: | :-------: | :-------: | :-------: | :-----: | :-----: |
| [rec_chinese_lite_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml) | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | | [rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml) | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc |
| [rec_chinese_common_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_common_train_v1.1.yml) | CRNN | ResNet34_vd | None | BiLSTM | ctc | | [rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml) | CRNN | ResNet34_vd | None | BiLSTM | ctc |
| rec_chinese_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | | rec_chinese_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc |
| rec_chinese_common_train.yml | CRNN | ResNet34_vd | None | BiLSTM | ctc | | rec_chinese_common_train.yml | CRNN | ResNet34_vd | None | BiLSTM | ctc |
| rec_icdar15_train.yml | CRNN | Mobilenet_v3 large 0.5 | None | BiLSTM | ctc | | rec_icdar15_train.yml | CRNN | Mobilenet_v3 large 0.5 | None | BiLSTM | ctc |
...@@ -210,39 +208,69 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t ...@@ -210,39 +208,69 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t
| rec_r34_vd_tps_bilstm_ctc.yml | STARNet | Resnet34_vd | tps | BiLSTM | ctc | | rec_r34_vd_tps_bilstm_ctc.yml | STARNet | Resnet34_vd | tps | BiLSTM | ctc |
| rec_r50fpn_vd_none_srn.yml | SRN | Resnet50_fpn_vd | None | rnn | srn | | rec_r50fpn_vd_none_srn.yml | SRN | Resnet50_fpn_vd | None | rnn | srn |
训练中文数据，推荐使用[rec_chinese_lite_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml)，如您希望尝试其他算法在中文数据集上的效果，请参考下列说明修改配置文件： 训练中文数据，推荐使用[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)，如您希望尝试其他算法在中文数据集上的效果，请参考下列说明修改配置文件：
以 `rec_mv3_none_none_ctc.yml` 为例： 以 `rec_chinese_lite_train_v2.0.yml` 为例：
``` ```
Global: Global:
... ...
# 修改 image_shape 以适应长文本 # 添加自定义字典，如修改字典请将路径指向新字典
image_shape: [3, 32, 320] character_dict_path: ppocr/utils/ppocr_keys_v1.txt
...
# 修改字符类型 # 修改字符类型
character_type: ch character_type: ch
# 添加自定义字典，如修改字典请将路径指向新字典
character_dict_path: ./ppocr/utils/ppocr_keys_v1.txt
# 训练时添加数据增强
distort: true
# 识别空格
use_space_char: true
...
# 修改reader类型
reader_yml: ./configs/rec/rec_chinese_reader.yml
... ...
# 识别空格
use_space_char: True
...
Optimizer: Optimizer:
... ...
# 添加学习率衰减策略 # 添加学习率衰减策略
decay: lr:
function: cosine_decay name: Cosine
# 每个 epoch 包含 iter 数 learning_rate: 0.001
step_each_epoch: 20 ...
# 总共训练epoch数
total_epoch: 1000 ...
Train:
dataset:
# 数据集格式，支持LMDBDateSet以及SimpleDataSet
name: SimpleDataSet
# 数据集路径
data_dir: ./train_data/
# 训练集标签文件
label_file_list: ["./train_data/train_list.txt"]
transforms:
...
- RecResizeImg:
# 修改 image_shape 以适应长文本
image_shape: [3, 32, 320]
...
loader:
...
# 单卡训练的batch_size
batch_size_per_card: 256
...
Eval:
dataset:
# 数据集格式，支持LMDBDateSet以及SimpleDataSet
name: SimpleDataSet
# 数据集路径
data_dir: ./train_data
# 验证集标签文件
label_file_list: ["./train_data/val_list.txt"]
transforms:
...
- RecResizeImg:
# 修改 image_shape 以适应长文本
image_shape: [3, 32, 320]
...
loader:
# 单卡验证的batch_size
batch_size_per_card: 256
...
``` ```
**注意，预测/评估时的配置文件请务必与训练一致。** **注意，预测/评估时的配置文件请务必与训练一致。**
...@@ -270,39 +298,41 @@ Global: ...@@ -270,39 +298,41 @@ Global:
... ...
# 添加自定义字典，如修改字典请将路径指向新字典 # 添加自定义字典，如修改字典请将路径指向新字典
character_dict_path: ./ppocr/utils/dict/french_dict.txt character_dict_path: ./ppocr/utils/dict/french_dict.txt
# 训练时添加数据增强
distort: true
# 识别空格
use_space_char: true
...
# 修改reader类型
reader_yml: ./configs/rec/multi_languages/rec_french_reader.yml
...
...
```
同时需要修改数据读取文件 `rec_french_reader.yml`：
```
TrainReader:
... ...
# 修改训练数据存放的目录名 # 识别空格
img_set_dir: ./train_data use_space_char: True
# 修改 label 文件名称
label_file_path: ./train_data/french_train.txt
... ...
Train:
dataset:
# 数据集格式，支持LMDBDateSet以及SimpleDataSet
name: SimpleDataSet
# 数据集路径
data_dir: ./train_data/
# 训练集标签文件
label_file_list: ["./train_data/french_train.txt"]
...
Eval:
dataset:
# 数据集格式，支持LMDBDateSet以及SimpleDataSet
name: SimpleDataSet
# 数据集路径
data_dir: ./train_data
# 验证集标签文件
label_file_list: ["./train_data/french_val.txt"]
...
``` ```
<a name="评估"></a> <a name="评估"></a>
### 评估 ### 评估
评估数据集可以通过 `configs/rec/rec_icdar15_reader.yml` 修改EvalReader中的 `label_file_path` 设置。 评估数据集可以通过 `configs/rec/rec_icdar15_train.yml` 修改Eval中的 `label_file_path` 设置。
*注意* 评估时必须确保配置文件中 infer_img 字段为空 *注意* 评估时必须确保配置文件中 infer_img 字段为空
``` ```
export CUDA_VISIBLE_DEVICES=0
# GPU 评估， Global.checkpoints 为待测权重 # GPU 评估， Global.checkpoints 为待测权重
python3 tools/eval.py -c configs/rec/rec_icdar15_train.yml -o Global.checkpoints={path/to/weights}/best_accuracy python3 -m paddle.distributed.launch --gpus '0' tools/eval.py -c configs/rec/rec_icdar15_train.yml -o Global.checkpoints={path/to/weights}/best_accuracy
``` ```
<a name="预测"></a> <a name="预测"></a>
...@@ -332,12 +362,12 @@ infer_img: doc/imgs_words/en/word_1.png ...@@ -332,12 +362,12 @@ infer_img: doc/imgs_words/en/word_1.png
word : joint word : joint
``` ```
预测使用的配置文件必须与训练一致，如您通过 `python3 tools/train.py -c configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml` 完成了中文模型的训练， 预测使用的配置文件必须与训练一致，如您通过 `python3 tools/train.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml` 完成了中文模型的训练，
您可以使用如下命令进行中文模型预测。 您可以使用如下命令进行中文模型预测。
``` ```
# 预测中文结果 # 预测中文结果
python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml -o Global.checkpoints={path/to/weights}/best_accuracy Global.infer_img=doc/imgs_words/ch/word_1.jpg python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml -o Global.checkpoints={path/to/weights}/best_accuracy Global.infer_img=doc/imgs_words/ch/word_1.jpg
``` ```
预测图片： 预测图片：
......
doc/doc_en/algorithm_overview_en.md
View file @ 90b968d5
...@@ -19,17 +19,17 @@ On the ICDAR2015 dataset, the text detection result is as follows: ...@@ -19,17 +19,17 @@ On the ICDAR2015 dataset, the text detection result is as follows:
|Model|Backbone|precision|recall|Hmean|Download link| |Model|Backbone|precision|recall|Hmean|Download link|
|-|-|-|-|-|-| |-|-|-|-|-|-|
|EAST|ResNet50_vd|88.18%|85.51%|86.82%|[Download link](https://paddleocr.bj.bcebos.com/det_r50_vd_east.tar)| |EAST|ResNet50_vd||||[Coming soon]()|
|EAST|MobileNetV3|81.67%|79.83%|80.74%|[Download link](https://paddleocr.bj.bcebos.com/det_mv3_east.tar)| |EAST|MobileNetV3||||[Coming soon]()|
|DB|ResNet50_vd|83.79%|80.65%|82.19%|[Download link](https://paddleocr.bj.bcebos.com/det_r50_vd_db.tar)| |DB|ResNet50_vd||||[Coming soon]()|
|DB|MobileNetV3|75.92%|73.18%|74.53%|[Download link](https://paddleocr.bj.bcebos.com/det_mv3_db.tar)| |DB|MobileNetV3||||[Coming soon]()|
|SAST|ResNet50_vd|92.18%|82.96%|87.33%|[Download link](https://paddleocr.bj.bcebos.com/SAST/sast_r50_vd_icdar2015.tar)| |SAST|ResNet50_vd||||[Coming soon]()|
On Total-Text dataset, the text detection result is as follows: On Total-Text dataset, the text detection result is as follows:
|Model|Backbone|precision|recall|Hmean|Download link| |Model|Backbone|precision|recall|Hmean|Download link|
|-|-|-|-|-|-| |-|-|-|-|-|-|
|SAST|ResNet50_vd|88.74%|79.80%|84.03%|[Download link](https://paddleocr.bj.bcebos.com/SAST/sast_r50_vd_total_text.tar)| |SAST|ResNet50_vd||||[Coming soon]()|
**Note：** Additional data, like icdar2013, icdar2017, COCO-Text, ArT, was added to the model training of SAST. Download English public dataset in organized format used by PaddleOCR from [Baidu Drive](https://pan.baidu.com/s/12cPnZcVuV1zn5DOd4mqjVw) (download code: 2bpi). **Note：** Additional data, like icdar2013, icdar2017, COCO-Text, ArT, was added to the model training of SAST. Download English public dataset in organized format used by PaddleOCR from [Baidu Drive](https://pan.baidu.com/s/12cPnZcVuV1zn5DOd4mqjVw) (download code: 2bpi).
...@@ -49,18 +49,12 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r ...@@ -49,18 +49,12 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r
|Model|Backbone|Avg Accuracy|Module combination|Download link| |Model|Backbone|Avg Accuracy|Module combination|Download link|
|-|-|-|-|-| |-|-|-|-|-|
|Rosetta|Resnet34_vd|80.24%|rec_r34_vd_none_none_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_r34_vd_none_none_ctc.tar)| |Rosetta|Resnet34_vd||rec_r34_vd_none_none_ctc|[Coming soon]()|
|Rosetta|MobileNetV3|78.16%|rec_mv3_none_none_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_mv3_none_none_ctc.tar)| |Rosetta|MobileNetV3||rec_mv3_none_none_ctc|[Coming soon]()|
|CRNN|Resnet34_vd|82.20%|rec_r34_vd_none_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_r34_vd_none_bilstm_ctc.tar)| |CRNN|Resnet34_vd||rec_r34_vd_none_bilstm_ctc|[Coming soon]()|
|CRNN|MobileNetV3|79.37%|rec_mv3_none_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_mv3_none_bilstm_ctc.tar)| |CRNN|MobileNetV3||rec_mv3_none_bilstm_ctc|[Coming soon]()|
|STAR-Net|Resnet34_vd|83.93%|rec_r34_vd_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_r34_vd_tps_bilstm_ctc.tar)| |STAR-Net|Resnet34_vd||rec_r34_vd_tps_bilstm_ctc|[Coming soon]()|
|STAR-Net|MobileNetV3|81.56%|rec_mv3_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/rec_mv3_tps_bilstm_ctc.tar)| |STAR-Net|MobileNetV3||rec_mv3_tps_bilstm_ctc|[Coming soon]()|
|RARE|Resnet34_vd|84.90%|rec_r34_vd_tps_bilstm_attn|[Download link](https://paddleocr.bj.bcebos.com/rec_r34_vd_tps_bilstm_attn.tar)|
|RARE|MobileNetV3|83.32%|rec_mv3_tps_bilstm_attn|[Download link](https://paddleocr.bj.bcebos.com/rec_mv3_tps_bilstm_attn.tar)|
|SRN|Resnet50_vd_fpn|88.33%|rec_r50fpn_vd_none_srn|[Download link](https://paddleocr.bj.bcebos.com/SRN/rec_r50fpn_vd_none_srn.tar)|
**Note：** SRN model uses data expansion method to expand the two training sets mentioned above, and the expanded data can be downloaded from [Baidu Drive](https://pan.baidu.com/s/1-HSZ-ZVdqBF2HaBZ5pRAKA) (download code: y3ry).
The average accuracy of the two-stage training in the original paper is 89.74%, and that of one stage training in paddleocr is 88.33%. Both pre-trained weights can be downloaded [here](https://paddleocr.bj.bcebos.com/SRN/rec_r50fpn_vd_none_srn.tar).
Please refer to the document for training guide and use of PaddleOCR text recognition algorithms [Text recognition model training/evaluation/prediction](./doc/doc_en/recognition_en.md) Please refer to the document for training guide and use of PaddleOCR text recognition algorithms [Text recognition model training/evaluation/prediction](./doc/doc_en/recognition_en.md)
doc/doc_en/recognition_en.md
View file @ 90b968d5
...@@ -135,7 +135,7 @@ If you need to customize dic file, please add character_dict_path field in confi ...@@ -135,7 +135,7 @@ If you need to customize dic file, please add character_dict_path field in confi
<a name="Add_space_category"></a> <a name="Add_space_category"></a>
- Add space category - Add space category
If you want to support the recognition of the `space` category, please set the `use_space_char` field in the yml file to `true`. If you want to support the recognition of the `space` category, please set the `use_space_char` field in the yml file to `True`.
**Note: use_space_char only takes effect when character_type=ch** **Note: use_space_char only takes effect when character_type=ch**
...@@ -158,10 +158,9 @@ tar -xf rec_mv3_none_bilstm_ctc.tar && rm -rf rec_mv3_none_bilstm_ctc.tar ...@@ -158,10 +158,9 @@ tar -xf rec_mv3_none_bilstm_ctc.tar && rm -rf rec_mv3_none_bilstm_ctc.tar
Start training: Start training:
``` ```
# GPU training Support single card and multi-card training, specify the card number through CUDA_VISIBLE_DEVICES # GPU training Support single card and multi-card training, specify the card number through --gpus
export CUDA_VISIBLE_DEVICES=0,1,2,3
# Training icdar15 English data and saving the log as train_rec.log # Training icdar15 English data and saving the log as train_rec.log
python3 tools/train.py -c configs/rec/rec_icdar15_train.yml 2>&1 | tee train_rec.log python3 -m paddle.distributed.launch --gpus '0,1,2,3' tools/train.py -c configs/rec/rec_icdar15_train.yml
``` ```
<a name="Data_Augmentation"></a> <a name="Data_Augmentation"></a>
- Data Augmentation - Data Augmentation
...@@ -184,8 +183,8 @@ If the evaluation set is large, the test will be time-consuming. It is recommend ...@@ -184,8 +183,8 @@ If the evaluation set is large, the test will be time-consuming. It is recommend
| Configuration file | Algorithm | backbone | trans | seq | pred | | Configuration file | Algorithm | backbone | trans | seq | pred |
| :--------: | :-------: | :-------: | :-------: | :-----: | :-----: | | :--------: | :-------: | :-------: | :-------: | :-----: | :-----: |
| [rec_chinese_lite_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml) | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | | [rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml) | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc |
| [rec_chinese_common_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_common_train_v1.1.yml) | CRNN | ResNet34_vd | None | BiLSTM | ctc | | [rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml) | CRNN | ResNet34_vd | None | BiLSTM | ctc |
| rec_chinese_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc | | rec_chinese_lite_train.yml | CRNN | Mobilenet_v3 small 0.5 | None | BiLSTM | ctc |
| rec_chinese_common_train.yml | CRNN | ResNet34_vd | None | BiLSTM | ctc | | rec_chinese_common_train.yml | CRNN | ResNet34_vd | None | BiLSTM | ctc |
| rec_icdar15_train.yml | CRNN | Mobilenet_v3 large 0.5 | None | BiLSTM | ctc | | rec_icdar15_train.yml | CRNN | Mobilenet_v3 large 0.5 | None | BiLSTM | ctc |
...@@ -199,39 +198,69 @@ If the evaluation set is large, the test will be time-consuming. It is recommend ...@@ -199,39 +198,69 @@ If the evaluation set is large, the test will be time-consuming. It is recommend
| rec_r34_vd_tps_bilstm_ctc.yml | STARNet | Resnet34_vd | tps | BiLSTM | ctc | | rec_r34_vd_tps_bilstm_ctc.yml | STARNet | Resnet34_vd | tps | BiLSTM | ctc |
For training Chinese data, it is recommended to use For training Chinese data, it is recommended to use
训练中文数据，推荐使用[rec_chinese_lite_train_v1.1.yml](../../configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml). If you want to try the result of other algorithms on the Chinese data set, please refer to the following instructions to modify the configuration file: [rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml). If you want to try the result of other algorithms on the Chinese data set, please refer to the following instructions to modify the configuration file:
co co
Take `rec_mv3_none_none_ctc.yml` as an example: Take `rec_chinese_lite_train_v2.0.yml` as an example:
``` ```
Global: Global:
... ...
# Modify image_shape to fit long text # Add a custom dictionary, such as modify the dictionary, please point the path to the new dictionary
image_shape: [3, 32, 320] character_dict_path: ppocr/utils/ppocr_keys_v1.txt
...
# Modify character type # Modify character type
character_type: ch character_type: ch
# Add a custom dictionary, such as modify the dictionary, please point the path to the new dictionary
character_dict_path: ./ppocr/utils/ppocr_keys_v1.txt
... ...
# Modify reader type
reader_yml: ./configs/rec/rec_chinese_reader.yml
# Whether to use data augmentation
distort: true
# Whether to recognize spaces # Whether to recognize spaces
use_space_char: true use_space_char: True
...
...
Optimizer: Optimizer:
... ...
# Add learning rate decay strategy # Add learning rate decay strategy
decay: lr:
function: cosine_decay name: Cosine
# Each epoch contains iter number learning_rate: 0.001
step_each_epoch: 20 ...
# Total epoch number
total_epoch: 1000 ...
Train:
dataset:
# Type of dataset，we support LMDBDateSet and SimpleDataSet
name: SimpleDataSet
# Path of dataset
data_dir: ./train_data/
# Path of train list
label_file_list: ["./train_data/train_list.txt"]
transforms:
...
- RecResizeImg:
# Modify image_shape to fit long text
image_shape: [3, 32, 320]
...
loader:
...
# Train batch_size for Single card
batch_size_per_card: 256
...
Eval:
dataset:
# Type of dataset，we support LMDBDateSet and SimpleDataSet
name: SimpleDataSet
# Path of dataset
data_dir: ./train_data
# Path of eval list
label_file_list: ["./train_data/val_list.txt"]
transforms:
...
- RecResizeImg:
# Modify image_shape to fit long text
image_shape: [3, 32, 320]
...
loader:
# Eval batch_size for Single card
batch_size_per_card: 256
...
``` ```
**Note that the configuration file for prediction/evaluation must be consistent with the training.** **Note that the configuration file for prediction/evaluation must be consistent with the training.**
...@@ -257,18 +286,33 @@ Take `rec_french_lite_train` as an example: ...@@ -257,18 +286,33 @@ Take `rec_french_lite_train` as an example:
``` ```
Global: Global:
... ...
# Add a custom dictionary, if you modify the dictionary # Add a custom dictionary, such as modify the dictionary, please point the path to the new dictionary
# please point the path to the new dictionary
character_dict_path: ./ppocr/utils/dict/french_dict.txt character_dict_path: ./ppocr/utils/dict/french_dict.txt
# Add data augmentation during training
distort: true
# Identify spaces
use_space_char: true
...
# Modify reader type
reader_yml: ./configs/rec/multi_languages/rec_french_reader.yml
... ...
# Whether to recognize spaces
use_space_char: True
... ...
Train:
dataset:
# Type of dataset，we support LMDBDateSet and SimpleDataSet
name: SimpleDataSet
# Path of dataset
data_dir: ./train_data/
# Path of train list
label_file_list: ["./train_data/french_train.txt"]
...
Eval:
dataset:
# Type of dataset，we support LMDBDateSet and SimpleDataSet
name: SimpleDataSet
# Path of dataset
data_dir: ./train_data
# Path of eval list
label_file_list: ["./train_data/french_val.txt"]
...
``` ```
<a name="EVALUATION"></a> <a name="EVALUATION"></a>
...@@ -277,9 +321,8 @@ Global: ...@@ -277,9 +321,8 @@ Global:
The evaluation data set can be modified via `configs/rec/rec_icdar15_reader.yml` setting of `label_file_path` in EvalReader. The evaluation data set can be modified via `configs/rec/rec_icdar15_reader.yml` setting of `label_file_path` in EvalReader.
``` ```
export CUDA_VISIBLE_DEVICES=0
# GPU evaluation, Global.checkpoints is the weight to be tested # GPU evaluation, Global.checkpoints is the weight to be tested
python3 tools/eval.py -c configs/rec/rec_icdar15_reader.yml -o Global.checkpoints={path/to/weights}/best_accuracy python3 -m paddle.distributed.launch --gpus '0' tools/eval.py -c configs/rec/rec_icdar15_reader.yml -o Global.checkpoints={path/to/weights}/best_accuracy
``` ```
<a name="PREDICTION"></a> <a name="PREDICTION"></a>
...@@ -294,7 +337,7 @@ The default prediction picture is stored in `infer_img`, and the weight is speci ...@@ -294,7 +337,7 @@ The default prediction picture is stored in `infer_img`, and the weight is speci
``` ```
# Predict English results # Predict English results
python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml -o Global.checkpoints={path/to/weights}/best_accuracy TestReader.infer_img=doc/imgs_words/en/word_1.jpg python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml -o Global.checkpoints={path/to/weights}/best_accuracy TestReader.infer_img=doc/imgs_words/en/word_1.jpg
``` ```
Input image: Input image:
...@@ -309,11 +352,11 @@ infer_img: doc/imgs_words/en/word_1.png ...@@ -309,11 +352,11 @@ infer_img: doc/imgs_words/en/word_1.png
word : joint word : joint
``` ```
The configuration file used for prediction must be consistent with the training. For example, you completed the training of the Chinese model with `python3 tools/train.py -c configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml`, you can use the following command to predict the Chinese model: The configuration file used for prediction must be consistent with the training. For example, you completed the training of the Chinese model with `python3 tools/train.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml`, you can use the following command to predict the Chinese model:
``` ```
# Predict Chinese results # Predict Chinese results
python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v1.1/rec_chinese_lite_train_v1.1.yml -o Global.checkpoints={path/to/weights}/best_accuracy TestReader.infer_img=doc/imgs_words/ch/word_1.jpg python3 tools/infer_rec.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml -o Global.checkpoints={path/to/weights}/best_accuracy TestReader.infer_img=doc/imgs_words/ch/word_1.jpg
``` ```
Input image: Input image:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment