Skip to content

GitLab

"doc/git@developer.sourcefind.cn:wangsen/paddle_dbnet.git" did not exist on "85aeae712ab3ec1193c6c503ae06352892dcd9df"
Commit 1d7c9f81 authored by LDOUBLEV's avatar LDOUBLEV
Browse files

Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into test_v11

parents 3f3b915b b1d26ded
Hide whitespace changes
Inline Side-by-side
Showing with 684 additions and 607 deletions
benchmark/run_benchmark_det.sh
View file @ 1d7c9f81
...@@ -17,7 +17,7 @@ function _set_params(){ ...@@ -17,7 +17,7 @@ function _set_params(){
skip_steps=2 # 解析日志，有些模型前几个step耗时长，需要跳过 (必填) skip_steps=2 # 解析日志，有些模型前几个step耗时长，需要跳过 (必填)
keyword="ips:" # 解析日志，筛选出数据所在行的关键字 (必填) keyword="ips:" # 解析日志，筛选出数据所在行的关键字 (必填)
index="1" index="1"
model_name=${model_item}_${run_mode}_bs${batch_size}_${fp_item} # model_item 用于yml文件名匹配，model_name 用于数据入库前端展示 model_name=${model_item}_bs${batch_size}_${fp_item} # model_item 用于yml文件名匹配，model_name 用于数据入库前端展示
# 以下不用修改 # 以下不用修改
device=${CUDA_VISIBLE_DEVICES//,/ } device=${CUDA_VISIBLE_DEVICES//,/ }
arr=(${device}) arr=(${device})
......
benchmark/run_det.sh
View file @ 1d7c9f81
...@@ -2,6 +2,7 @@ ...@@ -2,6 +2,7 @@
# 提供可稳定复现性能的脚本，默认在标准docker环境内py37执行： paddlepaddle/paddle:latest-gpu-cuda10.1-cudnn7 paddle=2.1.2 py=37 # 提供可稳定复现性能的脚本，默认在标准docker环境内py37执行： paddlepaddle/paddle:latest-gpu-cuda10.1-cudnn7 paddle=2.1.2 py=37
# 执行目录: ./PaddleOCR # 执行目录: ./PaddleOCR
# 1 安装该模型需要的依赖 (如需开启优化策略请注明) # 1 安装该模型需要的依赖 (如需开启优化策略请注明)
log_path=${LOG_PATH_INDEX_DIR:-$(pwd)}
python -m pip install -r requirements.txt python -m pip install -r requirements.txt
# 2 拷贝该模型需要数据、预训练模型 # 2 拷贝该模型需要数据、预训练模型
wget -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar && cd train_data && tar xf icdar2015.tar && cd ../ wget -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar && cd train_data && tar xf icdar2015.tar && cd ../
...@@ -12,18 +13,22 @@ wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dyg ...@@ -12,18 +13,22 @@ wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dyg
model_mode_list=(det_res18_db_v2.0 det_r50_vd_east det_r50_vd_pse) model_mode_list=(det_res18_db_v2.0 det_r50_vd_east det_r50_vd_pse)
fp_item_list=(fp32) fp_item_list=(fp32)
bs_list=(8 16)
for model_mode in ${model_mode_list[@]}; do for model_mode in ${model_mode_list[@]}; do
for fp_item in ${fp_item_list[@]}; do for fp_item in ${fp_item_list[@]}; do
if [ ${model_mode} == "det_r50_vd_east" ]; then
bs_list=(16)
else
bs_list=(8 16)
fi
for bs_item in ${bs_list[@]}; do for bs_item in ${bs_list[@]}; do
echo "index is speed, 1gpus, begin, ${model_name}" echo "index is speed, 1gpus, begin, ${model_name}"
run_mode=sp run_mode=sp
log_name=ocr_${model_mode}_${run_mode}_bs${bs_item}_${fp_item} log_name=ocr_${model_mode}_bs${bs_item}_${fp_item}_${run_mode}
CUDA_VISIBLE_DEVICES=0 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 1 ${model_mode} | tee ${log_path}/${log_name}_speed_1gpus 2>&1 # (5min) CUDA_VISIBLE_DEVICES=0 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 1 ${model_mode} | tee ${log_path}/${log_name}_speed_1gpus 2>&1 # (5min)
sleep 60 sleep 60
echo "index is speed, 8gpus, run_mode is multi_process, begin, ${model_name}" echo "index is speed, 8gpus, run_mode is multi_process, begin, ${model_name}"
run_mode=mp run_mode=mp
log_name=ocr_${model_mode}_${run_mode}_bs${bs_item}_${fp_item} log_name=ocr_${model_mode}_bs${bs_item}_${fp_item}_${run_mode}
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 2 ${model_mode} | tee ${log_path}/${log_name}_speed_8gpus8p 2>&1 CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 2 ${model_mode} | tee ${log_path}/${log_name}_speed_8gpus8p 2>&1
sleep 60 sleep 60
done done
......
deploy/slim/prune/export_prune_model.py
View file @ 1d7c9f81
...@@ -52,12 +52,17 @@ def main(config, device, logger, vdl_writer): ...@@ -52,12 +52,17 @@ def main(config, device, logger, vdl_writer):
config['Architecture']["Head"]['out_channels'] = char_num config['Architecture']["Head"]['out_channels'] = char_num
model = build_model(config['Architecture']) model = build_model(config['Architecture'])
flops = paddle.flops(model, [1, 3, 640, 640]) if config['Architecture']['model_type'] == 'det':
logger.info(f"FLOPs before pruning: {flops}") input_shape = [1, 3, 640, 640]
elif config['Architecture']['model_type'] == 'rec':
input_shape = [1, 3, 32, 320]
flops = paddle.flops(model, input_shape)
logger.info("FLOPs before pruning: {}".format(flops))
from paddleslim.dygraph import FPGMFilterPruner from paddleslim.dygraph import FPGMFilterPruner
model.train() model.train()
pruner = FPGMFilterPruner(model, [1, 3, 640, 640]) pruner = FPGMFilterPruner(model, input_shape)
# build metric # build metric
eval_class = build_metric(config['Metric']) eval_class = build_metric(config['Metric'])
...@@ -65,8 +70,13 @@ def main(config, device, logger, vdl_writer): ...@@ -65,8 +70,13 @@ def main(config, device, logger, vdl_writer):
def eval_fn(): def eval_fn():
metric = program.eval(model, valid_dataloader, post_process_class, metric = program.eval(model, valid_dataloader, post_process_class,
eval_class) eval_class)
logger.info(f"metric['hmean']: {metric['hmean']}") if config['Architecture']['model_type'] == 'det':
return metric['hmean'] main_indicator = 'hmean'
else:
main_indicator = 'acc'
logger.info("metric[{}]: {}".format(main_indicator, metric[
main_indicator]))
return metric[main_indicator]
params_sensitive = pruner.sensitive( params_sensitive = pruner.sensitive(
eval_func=eval_fn, eval_func=eval_fn,
...@@ -81,18 +91,22 @@ def main(config, device, logger, vdl_writer): ...@@ -81,18 +91,22 @@ def main(config, device, logger, vdl_writer):
# calculate pruned params's ratio # calculate pruned params's ratio
params_sensitive = pruner._get_ratios_by_loss(params_sensitive, loss=0.02) params_sensitive = pruner._get_ratios_by_loss(params_sensitive, loss=0.02)
for key in params_sensitive.keys(): for key in params_sensitive.keys():
logger.info(f"{key}, {params_sensitive[key]}") logger.info("{}, {}".format(key, params_sensitive[key]))
plan = pruner.prune_vars(params_sensitive, [0]) plan = pruner.prune_vars(params_sensitive, [0])
flops = paddle.flops(model, [1, 3, 640, 640]) flops = paddle.flops(model, input_shape)
logger.info(f"FLOPs after pruning: {flops}") logger.info("FLOPs after pruning: {}".format(flops))
# load pretrain model # load pretrain model
load_model(config, model) load_model(config, model)
metric = program.eval(model, valid_dataloader, post_process_class, metric = program.eval(model, valid_dataloader, post_process_class,
eval_class) eval_class)
logger.info(f"metric['hmean']: {metric['hmean']}") if config['Architecture']['model_type'] == 'det':
main_indicator = 'hmean'
else:
main_indicator = 'acc'
logger.info("metric['']: {}".format(main_indicator, metric[main_indicator]))
# start export model # start export model
from paddle.jit import to_static from paddle.jit import to_static
......
deploy/slim/prune/sensitivity_anal.py
View file @ 1d7c9f81
...@@ -73,13 +73,18 @@ def main(config, device, logger, vdl_writer): ...@@ -73,13 +73,18 @@ def main(config, device, logger, vdl_writer):
char_num = len(getattr(post_process_class, 'character')) char_num = len(getattr(post_process_class, 'character'))
config['Architecture']["Head"]['out_channels'] = char_num config['Architecture']["Head"]['out_channels'] = char_num
model = build_model(config['Architecture']) model = build_model(config['Architecture'])
if config['Architecture']['model_type'] == 'det':
input_shape = [1, 3, 640, 640]
elif config['Architecture']['model_type'] == 'rec':
input_shape = [1, 3, 32, 320]
flops = paddle.flops(model, input_shape)
flops = paddle.flops(model, [1, 3, 640, 640])
logger.info("FLOPs before pruning: {}".format(flops)) logger.info("FLOPs before pruning: {}".format(flops))
from paddleslim.dygraph import FPGMFilterPruner from paddleslim.dygraph import FPGMFilterPruner
model.train() model.train()
pruner = FPGMFilterPruner(model, [1, 3, 640, 640])
pruner = FPGMFilterPruner(model, input_shape)
# build loss # build loss
loss_class = build_loss(config['Loss']) loss_class = build_loss(config['Loss'])
...@@ -107,8 +112,14 @@ def main(config, device, logger, vdl_writer): ...@@ -107,8 +112,14 @@ def main(config, device, logger, vdl_writer):
def eval_fn(): def eval_fn():
metric = program.eval(model, valid_dataloader, post_process_class, metric = program.eval(model, valid_dataloader, post_process_class,
eval_class, False) eval_class, False)
logger.info("metric['hmean']: {}".format(metric['hmean'])) if config['Architecture']['model_type'] == 'det':
return metric['hmean'] main_indicator = 'hmean'
else:
main_indicator = 'acc'
logger.info("metric[{}]: {}".format(main_indicator, metric[
main_indicator]))
return metric[main_indicator]
run_sensitive_analysis = False run_sensitive_analysis = False
""" """
...@@ -149,7 +160,7 @@ def main(config, device, logger, vdl_writer): ...@@ -149,7 +160,7 @@ def main(config, device, logger, vdl_writer):
plan = pruner.prune_vars(params_sensitive, [0]) plan = pruner.prune_vars(params_sensitive, [0])
flops = paddle.flops(model, [1, 3, 640, 640]) flops = paddle.flops(model, input_shape)
logger.info("FLOPs after pruning: {}".format(flops)) logger.info("FLOPs after pruning: {}".format(flops))
# start train # start train
......
doc/doc_ch/code_and_doc.md 0 → 100644
View file @ 1d7c9f81
# 附录
本附录包含了Python、文档规范以及Pull Request流程，请各位开发者遵循相关内容
- [附录1：Python代码规范](#附录1)
- [附录2：文档规范](#附录2)
- [附录3：Pull Request说明](#附录3)
<a name="附录1"></a>
## 附录1：Python代码规范
PaddleOCR的Python代码遵循 [PEP8规范](https://www.python.org/dev/peps/pep-0008/)，其中一些关注的重点包括如下内容
- 空格
- 空格应该加在逗号、分号、冒号前，而非他们的后面
```python
# 正确：
print(x, y)
# 错误：
print(x , y)
```
- 在函数中指定关键字参数或默认参数值时, 不要在其两侧使用空格
```python
# 正确：
def complex(real, imag=0.0)
# 错误：
def complex(real, imag = 0.0)
```
- 注释
- 行内注释：行内注释使用 `#` 号表示，在代码与 `#` 之间需要空两个空格， `#` 与注释之间应当空一个空格，例如
```python
x = x + 1 # Compensate for border
```
- 函数和方法：每个函数的定义后的描述应该包括以下内容：
- 函数描述：函数的作用，输入输出的
- Args：每个参数的名字以及对该参数的描述
- Returns：返回值的含义和类型
```python
def fetch_bigtable_rows(big_table, keys, other_silly_variable=None):
"""Fetches rows from a Bigtable.
Retrieves rows pertaining to the given keys from the Table instance
represented by big_table. Silly things may happen if
other_silly_variable is not None.
Args:
big_table: An open Bigtable Table instance.
keys: A sequence of strings representing the key of each table row
to fetch.
other_silly_variable: Another optional variable, that has a much
longer name than the other args, and which does nothing.
Returns:
A dict mapping keys to the corresponding table row data
fetched. Each row is represented as a tuple of strings. For
example:
{'Serak': ('Rigel VII', 'Preparer'),
'Zim': ('Irk', 'Invader'),
'Lrrr': ('Omicron Persei 8', 'Emperor')}
If a key from the keys argument is missing from the dictionary,
then that row was not found in the table.
"""
pass
```
<a name="附录2"></a>
## 附录2：文档规范
### 2.1 总体说明
- 文档位置：如果您增加的新功能可以补充在原有的Markdown文件中，请**不要重新新建**一个文件。如果您对添加的位置不清楚，可以先PR代码，然后在commit中询问官方人员。
- 新增Markdown文档名称：使用英文描述文档内容，一般由小写字母与下划线组合而成，例如 `add_new_algorithm.md`
- 新增Markdown文档格式：目录 - 正文 - FAQ
> 目录生成方法可以使用 [此网站](https://ecotrust-canada.github.io/markdown-toc/) 将md内容复制之后自动提取目录，然后在md文件的每个标题前添加 `<a name="XXXX"></a>`
- 中英双语：任何对文档的改动或新增都需要分别在中文和英文文档上进行。
### 2.2 格式规范
- 标题格式：文档标题格式按照：阿拉伯数字小数点组合 - 空格 - 标题的格式（例如 `2.1 XXXX` ， `2. XXXX`）
- 代码块：通过代码块格式展示需要运行的代码，在代码块前描述命令参数的含义。例如：
> 检测+方向分类器+识别全流程：设置方向分类器参数 `--use_angle_cls true` 后可对竖排文本进行识别。
>
> ```
> paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true
> ```
- 变量引用：如果在行内引用到代码变量或命令参数，需要用行内代码表示，例如上方 `--use_angle_cls true` ，并在前后各空一格
- 补充说明：通过引用格式 `>` 补充说明，或对注意事项进行说明
- 图片：如果在说明文档中增加了图片，请规范图片的命名形式（描述图片内容），并将图片添加在 `doc/` 下
<a name="附录3"></a>
## 附录3：Pull Request说明
### 3.1 PaddleOCR分支说明
PaddleOCR未来将维护2种分支，分别为：
- release/x.x系列分支：为稳定的发行版本分支，也是默认分支。PaddleOCR会根据功能更新情况发布新的release分支，同时适配Paddle的release版本。随着版本迭代，release/x.x系列分支会越来越多，默认维护最新版本的release分支。
- dygraph分支：为开发分支，适配Paddle动态图的dygraph版本，主要用于开发新功能。如果有同学需要进行二次开发，请选择dygraph分支。为了保证dygraph分支能在需要的时候拉出release/x.x分支，dygraph分支的代码只能使用Paddle最新release分支中有效的api。也就是说，如果Paddle dygraph分支中开发了新的api，但尚未出现在release分支代码中，那么请不要在PaddleOCR中使用。除此之外，对于不涉及api的性能优化、参数调整、策略更新等，都可以正常进行开发。
PaddleOCR的历史分支，未来将不再维护。考虑到一些同学可能仍在使用，这些分支还会继续保留：
- develop分支：这个分支曾用于静态图的开发与测试，目前兼容>=1.7版本的Paddle。如果有特殊需求，要适配旧版本的Paddle，那还可以使用这个分支，但除了修复bug外不再更新代码。
PaddleOCR欢迎大家向repo中积极贡献代码，下面给出一些贡献代码的基本流程。
### 3.2 PaddleOCR代码提交流程与规范
> 如果你熟悉Git使用，可以直接跳转到 [3.2.10 提交代码的一些约定](#提交代码的一些约定)
#### 3.2.1 创建你的 `远程仓库`
- 在PaddleOCR的 [GitHub首页](https://github.com/PaddlePaddle/PaddleOCR)，点击左上角 `Fork` 按钮，在你的个人目录下创建 `远程仓库`，比如`https://github.com/{your_name}/PaddleOCR`。
![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/banner.png)
- 将 `远程仓库` Clone到本地
```
# 拉取develop分支的代码
git clone https://github.com/{your_name}/PaddleOCR.git -b dygraph
cd PaddleOCR
```
> 多数情况下clone失败是由于网络原因，请稍后重试或配置代理
#### 3.2.2 和 `远程仓库` 建立连接
首先查看当前 `远程仓库` 的信息。
```
git remote -v
# origin https://github.com/{your_name}/PaddleOCR.git (fetch)
# origin https://github.com/{your_name}/PaddleOCR.git (push)
```
只有clone的 `远程仓库` 的信息，也就是自己用户名下的 PaddleOCR，接下来我们创建一个原始 PaddleOCR 仓库的远程主机，命名为 upstream。
```
git remote add upstream https://github.com/PaddlePaddle/PaddleOCR.git
```
使用 `git remote -v` 查看当前 `远程仓库` 的信息，输出如下，发现包括了origin和upstream 2个 `远程仓库` 。
```
origin https://github.com/{your_name}/PaddleOCR.git (fetch)
origin https://github.com/{your_name}/PaddleOCR.git (push)
upstream https://github.com/PaddlePaddle/PaddleOCR.git (fetch)
upstream https://github.com/PaddlePaddle/PaddleOCR.git (push)
```
这主要是为了后续在提交pull request(PR)时，始终保持本地仓库最新。
#### 3.2.3 创建本地分支
可以基于当前分支创建新的本地分支，命令如下。
```
git checkout -b new_branch
```
也可以基于远程或者上游的分支创建新的分支，命令如下。
```
# 基于用户远程仓库(origin)的develop创建new_branch分支
git checkout -b new_branch origin/develop
# 基于上游远程仓库(upstream)的develop创建new_branch分支
# 如果需要从upstream创建新的分支，需要首先使用git fetch upstream获取上游代码
git checkout -b new_branch upstream/develop
```
最终会显示切换到新的分支，输出信息如下
```
Branch new_branch set up to track remote branch develop from upstream.
Switched to a new branch 'new_branch'
```
#### 3.2.4 使用pre-commit勾子
Paddle 开发人员使用 pre-commit 工具来管理 Git 预提交钩子。 它可以帮助我们格式化源代码（C++，Python），在提交（commit）前自动检查一些基本事宜（如每个文件只有一个 EOL，Git 中不要添加大文件等）。
pre-commit测试是 Travis-CI 中单元测试的一部分，不满足钩子的 PR 不能被提交到 PaddleOCR，首先安装并在当前目录运行它：
```
pip install pre-commit
pre-commit install
```
> 1. Paddle 使用 clang-format 来调整 C/C++ 源代码格式，请确保 `clang-format` 版本在 3.8 以上。
>
> 2. 通过pip install pre-commit和conda install -c conda-forge pre-commit安装的yapf稍有不同的，PaddleOCR 开发人员使用的是 `pip install pre-commit`。
#### 3.2.5 修改与提交代码
假设对PaddleOCR的 `README.md` 做了一些修改，可以通过 `git status` 查看改动的文件，然后使用 `git add` 添加改动文件。
```
git status # 查看改动文件
git add README.md
pre-commit
```
重复上述步骤，直到pre-comit格式检查不报错。如下所示。
[![img](https://github.com/PaddlePaddle/PaddleClas/raw/release/2.3/docs/images/quick_start/community/003_precommit_pass.png)](https://github.com/PaddlePaddle/PaddleClas/blob/release/2.3/docs/images/quick_start/community/003_precommit_pass.png)
使用下面的命令完成提交。
```
git commit -m "your commit info"
```
#### 3.2.6 保持本地仓库最新
获取 upstream 的最新代码并更新当前分支。这里的upstream来自于2.2节的`和远程仓库建立连接`部分。
```
git fetch upstream
# 如果是希望提交到其他分支，则需要从upstream的其他分支pull代码，这里是develop
git pull upstream develop
```
#### 3.2.7 push到远程仓库
```
git push origin new_branch
```
#### 3.2.7 提交Pull Request
点击new pull request，选择本地分支和目标分支，如下图所示。在PR的描述说明中，填写该PR所完成的功能。接下来等待review，如果有需要修改的地方，参照上述步骤更新 origin 中的对应分支即可。
![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/pr.png)
#### 3.2.8 签署CLA协议和通过单元测试
- 签署CLA 在首次向PaddlePaddle提交Pull Request时，您需要您签署一次CLA(Contributor License Agreement)协议，以保证您的代码可以被合入，具体签署方式如下：
1. 请您查看PR中的Check部分，找到license/cla，并点击右侧detail，进入CLA网站
2. 点击CLA网站中的“Sign in with GitHub to agree”,点击完成后将会跳转回您的Pull Request页面
#### 3.2.9 删除分支
- 删除远程分支
在 PR 被 merge 进主仓库后，我们可以在 PR 的页面删除远程仓库的分支。
也可以使用 `git push origin :分支名` 删除远程分支，如：
```
git push origin :new_branch
```
- 删除本地分支
```
# 切换到develop分支，否则无法删除当前分支
git checkout develop
# 删除new_branch分支
git branch -D new_branch
```
<a name="提交代码的一些约定"></a>
#### 3.2.10 提交代码的一些约定
为了使官方维护人员在评审代码时更好地专注于代码本身，请您每次提交代码时，遵守以下约定：
1）请保证Travis-CI 中单元测试能顺利通过。如果没过，说明提交的代码存在问题，官方维护人员一般不做评审。
2）提交Pull Request前：
- 请注意commit的数量。
原因：如果仅仅修改一个文件但提交了十几个commit，每个commit只做了少量的修改，这会给评审人带来很大困扰。评审人需要逐一查看每个commit才能知道做了哪些修改，且不排除commit之间的修改存在相互覆盖的情况。
建议：每次提交时，保持尽量少的commit，可以通过git commit --amend补充上次的commit。对已经Push到远程仓库的多个commit，可以参考[squash commits after push](https://stackoverflow.com/questions/5667884/how-to-squash-commits-in-git-after-they-have-been-pushed)。
- 请注意每个commit的名称：应能反映当前commit的内容，不能太随意。
3）如果解决了某个Issue的问题，请在该Pull Request的第一个评论框中加上：fix #issue_number，这样当该Pull Request被合并后，会自动关闭对应的Issue。关键词包括：close, closes, closed, fix, fixes, fixed, resolve, resolves, resolved，请选择合适的词汇。详细可参考[Closing issues via commit messages](https://help.github.com/articles/closing-issues-via-commit-messages)。
此外，在回复评审人意见时，请您遵守以下约定：
1）官方维护人员的每一个review意见都希望得到回复，这样会更好地提升开源社区的贡献。
- 对评审意见同意且按其修改完的，给个简单的Done即可；
- 对评审意见不同意的，请给出您自己的反驳理由。
2）如果评审意见比较多:
- 请给出总体的修改情况。
- 请采用`start a review`进行回复，而非直接回复的方式。原因是每个回复都会发送一封邮件，会造成邮件灾难。
\ No newline at end of file
doc/doc_ch/detection.md
View file @ 1d7c9f81
...@@ -247,3 +247,7 @@ Q1: 训练模型转inference 模型之后预测效果不一致？ ...@@ -247,3 +247,7 @@ Q1: 训练模型转inference 模型之后预测效果不一致？
**A**：此类问题出现较多，问题多是trained model预测时候的预处理、后处理参数和inference model预测的时候的预处理、后处理参数不一致导致的。以det_mv3_db.yml配置文件训练的模型为例，训练模型、inference模型预测结果不一致问题解决方式如下： **A**：此类问题出现较多，问题多是trained model预测时候的预处理、后处理参数和inference model预测的时候的预处理、后处理参数不一致导致的。以det_mv3_db.yml配置文件训练的模型为例，训练模型、inference模型预测结果不一致问题解决方式如下：
- 检查[trained model预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L116)，和[inference model的预测预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/predict_det.py#L42)函数是否一致。算法在评估的时候，输入图像大小会影响精度，为了和论文保持一致，训练icdar15配置文件中将图像resize到[736, 1280]，但是在inference model预测的时候只有一套默认参数，会考虑到预测速度问题，默认限制图像最长边为960做resize的。训练模型预处理和inference模型的预处理函数位于[ppocr/data/imaug/operators.py](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/ppocr/data/imaug/operators.py#L147) - 检查[trained model预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L116)，和[inference model的预测预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/predict_det.py#L42)函数是否一致。算法在评估的时候，输入图像大小会影响精度，为了和论文保持一致，训练icdar15配置文件中将图像resize到[736, 1280]，但是在inference model预测的时候只有一套默认参数，会考虑到预测速度问题，默认限制图像最长边为960做resize的。训练模型预处理和inference模型的预处理函数位于[ppocr/data/imaug/operators.py](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/ppocr/data/imaug/operators.py#L147)
- 检查[trained model后处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L51)，和[inference 后处理参数](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/utility.py#L50)是否一致。 - 检查[trained model后处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L51)，和[inference 后处理参数](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/utility.py#L50)是否一致。
Q1: 训练EAST模型提示找不到lanms库？
**A**：执行pip3 install lanms-nova 即可。
doc/doc_ch/inference.md
View file @ 1d7c9f81
...@@ -34,6 +34,8 @@ inference 模型（`paddle.jit.save`保存的模型） ...@@ -34,6 +34,8 @@ inference 模型（`paddle.jit.save`保存的模型）
- [1. 超轻量中文OCR模型推理](#超轻量中文OCR模型推理) - [1. 超轻量中文OCR模型推理](#超轻量中文OCR模型推理)
- [2. 其他模型推理](#其他模型推理) - [2. 其他模型推理](#其他模型推理)
- [六、参数解释](参数解释)
<a name="训练模型转inference模型"></a> <a name="训练模型转inference模型"></a>
## 一、训练模型转inference模型 ## 一、训练模型转inference模型
...@@ -394,3 +396,127 @@ python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/img_10.jpg" --d ...@@ -394,3 +396,127 @@ python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/img_10.jpg" --d
执行命令后，识别结果图像如下： 执行命令后，识别结果图像如下：
![](../imgs_results/img_10_east_starnet.jpg) ![](../imgs_results/img_10_east_starnet.jpg)
<a name="参数解释"></a>
# 六、参数解释
更多关于预测过程的参数解释如下所示。
* 全局信息
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| image_dir | str | 无，必须显式指定 | 图像或者文件夹路径 |
| vis_font_path | str | "./doc/fonts/simfang.ttf" | 用于可视化的字体路径 |
| drop_score | float | 0.5 | 识别得分小于该值的结果会被丢弃，不会作为返回结果 |
| use_pdserving | bool | False | 是否使用Paddle Serving进行预测 |
| warmup | bool | False | 是否开启warmup，在统计预测耗时的时候，可以使用这种方法 |
| draw_img_save_dir | str | "./inference_results" | 系统串联预测OCR结果的保存文件夹 |
| save_crop_res | bool | False | 是否保存OCR的识别文本图像 |
| crop_res_save_dir | str | "./output" | 保存OCR识别出来的文本图像路径 |
| use_mp | bool | False | 是否开启多进程预测 |
| total_process_num | int | 6 | 开启的进城数，`use_mp`为`True`时生效 |
| process_id | int | 0 | 当前进程的id号，无需自己修改 |
| benchmark | bool | False | 是否开启benchmark，对预测速度、显存占用等进行统计 |
| save_log_path | str | "./log_output/" | 开启`benchmark`时，日志结果的保存文件夹 |
| show_log | bool | True | 是否显示预测中的日志信息 |
| use_onnx | bool | False | 是否开启onnx预测 |
* 预测引擎相关
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| use_gpu | bool | True | 是否使用GPU进行预测 |
| ir_optim | bool | True | 是否对计算图进行分析与优化，开启后可以加速预测过程 |
| use_tensorrt | bool | False | 是否开启tensorrt |
| min_subgraph_size | int | 15 | tensorrt中最小子图size，当子图的size大于该值时，才会尝试对该子图使用trt engine计算 |
| precision | str | fp32 | 预测的精度，支持`fp32`, `fp16`, `int8` 3种输入 |
| enable_mkldnn | bool | True | 是否开启mkldnn |
| cpu_threads | int | 10 | 开启mkldnn时，cpu预测的线程数 |
* 文本检测模型相关
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| det_algorithm | str | "DB" | 文本检测算法名称，目前支持`DB`, `EAST`, `SAST`, `PSE` |
| det_model_dir | str | xx | 检测inference模型路径 |
| det_limit_side_len | int | 960 | 检测的图像边长限制 |
| det_limit_type | str | "max" | 检测的变成限制类型，目前支持`min`, `max`，`min`表示保证图像最短边不小于`det_limit_side_len`，`max`表示保证图像最长边不大于`det_limit_side_len` |
其中，DB算法相关参数如下
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| det_db_thresh | float | 0.3 | DB输出的概率图中，得分大于该阈值的像素点才会被认为是文字像素点 |
| det_db_box_thresh | float | 0.6 | 检测结果边框内，所有像素点的平均得分大于该阈值时，该结果会被认为是文字区域 |
| det_db_unclip_ratio | float | 1.5 | `Vatti clipping`算法的扩张系数，使用该方法对文字区域进行扩张 |
| max_batch_size | int | 10 | 预测的batch size |
| use_dilation | bool | False | 是否对分割结果进行膨胀以获取更优检测效果 |
| det_db_score_mode | str | "fast" | DB的检测结果得分计算方法，支持`fast`和`slow`，`fast`是根据polygon的外接矩形边框内的所有像素计算平均得分，`slow`是根据原始polygon内的所有像素计算平均得分，计算速度相对较慢一些，但是更加准确一些。 |
EAST算法相关参数如下
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| det_east_score_thresh | float | 0.8 | EAST后处理中score map的阈值 |
| det_east_cover_thresh | float | 0.1 | EAST后处理中文本框的平均得分阈值 |
| det_east_nms_thresh | float | 0.2 | EAST后处理中nms的阈值 |
SAST算法相关参数如下
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| det_sast_score_thresh | float | 0.5 | SAST后处理中的得分阈值 |
| det_sast_nms_thresh | float | 0.5 | SAST后处理中nms的阈值 |
| det_sast_polygon | bool | False | 是否多边形检测，弯曲文本场景（如Total-Text）设置为True |
PSE算法相关参数如下
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| det_pse_thresh | float | 0.0 | 对输出图做二值化的阈值 |
| det_pse_box_thresh | float | 0.85 | 对box进行过滤的阈值，低于此阈值的丢弃 |
| det_pse_min_area | float | 16 | box的最小面积，低于此阈值的丢弃 |
| det_pse_box_type | str | "box" | 返回框的类型，box:四点坐标，poly: 弯曲文本的所有点坐标 |
| det_pse_scale | int | 1 | 输入图像相对于进后处理的图的比例，如`640*640`的图像，网络输出为`160*160`，scale为2的情况下，进后处理的图片shape为`320*320`。这个值调大可以加快后处理速度，但是会带来精度的下降 |
* 文本识别模型相关
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| rec_algorithm | str | "CRNN" | 文本识别算法名称，目前支持`CRNN`, `SRN`, `RARE`, `NETR`, `SAR` |
| rec_model_dir | str | 无，如果使用识别模型，该项是必填项 | 识别inference模型路径 |
| rec_image_shape | list | [3, 32, 320] | 识别时的图像尺寸， |
| rec_batch_num | int | 6 | 识别的batch size |
| max_text_length | int | 25 | 识别结果最大长度，在`SRN`中有效 |
| rec_char_dict_path | str | "./ppocr/utils/ppocr_keys_v1.txt" | 识别的字符字典文件 |
| use_space_char | bool | True | 是否包含空格，如果为`True`，则会在最后字符字典中补充`空格`字符 |
* 端到端文本检测与识别模型相关
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| e2e_algorithm | str | "PGNet" | 端到端算法名称，目前支持`PGNet` |
| e2e_model_dir | str | 无，如果使用端到端模型，该项是必填项 | 端到端模型inference模型路径 |
| e2e_limit_side_len | int | 768 | 端到端的输入图像边长限制 |
| e2e_limit_type | str | "max" | 端到端的边长限制类型，目前支持`min`, `max`，`min`表示保证图像最短边不小于`e2e_limit_side_len`，`max`表示保证图像最长边不大于`e2e_limit_side_len` |
| e2e_pgnet_score_thresh | float | xx | xx |
| e2e_char_dict_path | str | "./ppocr/utils/ic15_dict.txt" | 识别的字典文件路径 |
| e2e_pgnet_valid_set | str | "totaltext" | 验证集名称，目前支持`totaltext`, `partvgg`，不同数据集对应的后处理方式不同，与训练过程保持一致即可 |
| e2e_pgnet_mode | str | "fast" | PGNet的检测结果得分计算方法，支持`fast`和`slow`，`fast`是根据polygon的外接矩形边框内的所有像素计算平均得分，`slow`是根据原始polygon内的所有像素计算平均得分，计算速度相对较慢一些，但是更加准确一些。 |
* 方向分类器模型相关
| 参数名称 | 类型 | 默认值 | 含义 |
| :--: | :--: | :--: | :--: |
| use_angle_cls | bool | False | 是否使用方向分类器 |
| cls_model_dir | str | 无，如果需要使用，则必须显式指定路径 | 方向分类器inference模型路径 |
| cls_image_shape | list | [3, 48, 192] | 预测尺度 |
| label_list | list | ['0', '180'] | class id对应的角度值 |
| cls_batch_num | int | 6 | 方向分类器预测的batch size |
| cls_thresh | float | 0.9 | 预测阈值，模型预测结果为180度，且得分大于该阈值时，认为最终预测结果为180度，需要翻转 |
doc/joinus.PNG
View replaced file @ 3f3b915b View file @ 1d7c9f81
doc/joinus.PNG

193 KB | W: | H:

doc/joinus.PNG

193 KB | W: | H:

doc/joinus.PNG
doc/joinus.PNG
doc/joinus.PNG
doc/joinus.PNG
  • 2-up
  • Swipe
  • Onion skin
ppocr/modeling/backbones/__init__.py
View file @ 1d7c9f81
...@@ -16,7 +16,7 @@ __all__ = ["build_backbone"] ...@@ -16,7 +16,7 @@ __all__ = ["build_backbone"]
def build_backbone(config, model_type): def build_backbone(config, model_type):
if model_type == "det": if model_type == "det" or model_type == "table":
from .det_mobilenet_v3 import MobileNetV3 from .det_mobilenet_v3 import MobileNetV3
from .det_resnet_vd import ResNet from .det_resnet_vd import ResNet
from .det_resnet_vd_sast import ResNet_SAST from .det_resnet_vd_sast import ResNet_SAST
...@@ -36,10 +36,6 @@ def build_backbone(config, model_type): ...@@ -36,10 +36,6 @@ def build_backbone(config, model_type):
elif model_type == "e2e": elif model_type == "e2e":
from .e2e_resnet_vd_pg import ResNet from .e2e_resnet_vd_pg import ResNet
support_dict = ["ResNet"] support_dict = ["ResNet"]
elif model_type == "table":
from .table_resnet_vd import ResNet
from .table_mobilenet_v3 import MobileNetV3
support_dict = ["ResNet", "MobileNetV3"]
else: else:
raise NotImplementedError raise NotImplementedError
......
ppocr/modeling/backbones/rec_mobilenet_v3.py
View file @ 1d7c9f81
...@@ -26,8 +26,10 @@ class MobileNetV3(nn.Layer): ...@@ -26,8 +26,10 @@ class MobileNetV3(nn.Layer):
scale=0.5, scale=0.5,
large_stride=None, large_stride=None,
small_stride=None, small_stride=None,
disable_se=False,
**kwargs): **kwargs):
super(MobileNetV3, self).__init__() super(MobileNetV3, self).__init__()
self.disable_se = disable_se
if small_stride is None: if small_stride is None:
small_stride = [2, 2, 2, 2] small_stride = [2, 2, 2, 2]
if large_stride is None: if large_stride is None:
...@@ -101,6 +103,7 @@ class MobileNetV3(nn.Layer): ...@@ -101,6 +103,7 @@ class MobileNetV3(nn.Layer):
block_list = [] block_list = []
inplanes = make_divisible(inplanes * scale) inplanes = make_divisible(inplanes * scale)
for (k, exp, c, se, nl, s) in cfg: for (k, exp, c, se, nl, s) in cfg:
se = se and not self.disable_se
block_list.append( block_list.append(
ResidualUnit( ResidualUnit(
in_channels=inplanes, in_channels=inplanes,
......
ppocr/modeling/backbones/table_mobilenet_v3.py deleted 100644 → 0
View file @ 3f3b915b
# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import paddle
from paddle import nn
import paddle.nn.functional as F
from paddle import ParamAttr
__all__ = ['MobileNetV3']
def make_divisible(v, divisor=8, min_value=None):
if min_value is None:
min_value = divisor
new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
if new_v < 0.9 * v:
new_v += divisor
return new_v
class MobileNetV3(nn.Layer):
def __init__(self,
in_channels=3,
model_name='large',
scale=0.5,
disable_se=False,
**kwargs):
"""
the MobilenetV3 backbone network for detection module.
Args:
params(dict): the super parameters for build network
"""
super(MobileNetV3, self).__init__()
self.disable_se = disable_se
if model_name == "large":
cfg = [
# k, exp, c, se, nl, s,
[3, 16, 16, False, 'relu', 1],
[3, 64, 24, False, 'relu', 2],
[3, 72, 24, False, 'relu', 1],
[5, 72, 40, True, 'relu', 2],
[5, 120, 40, True, 'relu', 1],
[5, 120, 40, True, 'relu', 1],
[3, 240, 80, False, 'hardswish', 2],
[3, 200, 80, False, 'hardswish', 1],
[3, 184, 80, False, 'hardswish', 1],
[3, 184, 80, False, 'hardswish', 1],
[3, 480, 112, True, 'hardswish', 1],
[3, 672, 112, True, 'hardswish', 1],
[5, 672, 160, True, 'hardswish', 2],
[5, 960, 160, True, 'hardswish', 1],
[5, 960, 160, True, 'hardswish', 1],
]
cls_ch_squeeze = 960
elif model_name == "small":
cfg = [
# k, exp, c, se, nl, s,
[3, 16, 16, True, 'relu', 2],
[3, 72, 24, False, 'relu', 2],
[3, 88, 24, False, 'relu', 1],
[5, 96, 40, True, 'hardswish', 2],
[5, 240, 40, True, 'hardswish', 1],
[5, 240, 40, True, 'hardswish', 1],
[5, 120, 48, True, 'hardswish', 1],
[5, 144, 48, True, 'hardswish', 1],
[5, 288, 96, True, 'hardswish', 2],
[5, 576, 96, True, 'hardswish', 1],
[5, 576, 96, True, 'hardswish', 1],
]
cls_ch_squeeze = 576
else:
raise NotImplementedError("mode[" + model_name +
"_model] is not implemented!")
supported_scale = [0.35, 0.5, 0.75, 1.0, 1.25]
assert scale in supported_scale, \
"supported scale are {} but input scale is {}".format(supported_scale, scale)
inplanes = 16
# conv1
self.conv = ConvBNLayer(
in_channels=in_channels,
out_channels=make_divisible(inplanes * scale),
kernel_size=3,
stride=2,
padding=1,
groups=1,
if_act=True,
act='hardswish',
name='conv1')
self.stages = []
self.out_channels = []
block_list = []
i = 0
inplanes = make_divisible(inplanes * scale)
for (k, exp, c, se, nl, s) in cfg:
se = se and not self.disable_se
start_idx = 2 if model_name == 'large' else 0
if s == 2 and i > start_idx:
self.out_channels.append(inplanes)
self.stages.append(nn.Sequential(*block_list))
block_list = []
block_list.append(
ResidualUnit(
in_channels=inplanes,
mid_channels=make_divisible(scale * exp),
out_channels=make_divisible(scale * c),
kernel_size=k,
stride=s,
use_se=se,
act=nl,
name="conv" + str(i + 2)))
inplanes = make_divisible(scale * c)
i += 1
block_list.append(
ConvBNLayer(
in_channels=inplanes,
out_channels=make_divisible(scale * cls_ch_squeeze),
kernel_size=1,
stride=1,
padding=0,
groups=1,
if_act=True,
act='hardswish',
name='conv_last'))
self.stages.append(nn.Sequential(*block_list))
self.out_channels.append(make_divisible(scale * cls_ch_squeeze))
for i, stage in enumerate(self.stages):
self.add_sublayer(sublayer=stage, name="stage{}".format(i))
def forward(self, x):
x = self.conv(x)
out_list = []
for stage in self.stages:
x = stage(x)
out_list.append(x)
return out_list
class ConvBNLayer(nn.Layer):
def __init__(self,
in_channels,
out_channels,
kernel_size,
stride,
padding,
groups=1,
if_act=True,
act=None,
name=None):
super(ConvBNLayer, self).__init__()
self.if_act = if_act
self.act = act
self.conv = nn.Conv2D(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=kernel_size,
stride=stride,
padding=padding,
groups=groups,
weight_attr=ParamAttr(name=name + '_weights'),
bias_attr=False)
self.bn = nn.BatchNorm(
num_channels=out_channels,
act=None,
param_attr=ParamAttr(name=name + "_bn_scale"),
bias_attr=ParamAttr(name=name + "_bn_offset"),
moving_mean_name=name + "_bn_mean",
moving_variance_name=name + "_bn_variance")
def forward(self, x):
x = self.conv(x)
x = self.bn(x)
if self.if_act:
if self.act == "relu":
x = F.relu(x)
elif self.act == "hardswish":
x = F.hardswish(x)
else:
print("The activation function({}) is selected incorrectly.".
format(self.act))
exit()
return x
class ResidualUnit(nn.Layer):
def __init__(self,
in_channels,
mid_channels,
out_channels,
kernel_size,
stride,
use_se,
act=None,
name=''):
super(ResidualUnit, self).__init__()
self.if_shortcut = stride == 1 and in_channels == out_channels
self.if_se = use_se
self.expand_conv = ConvBNLayer(
in_channels=in_channels,
out_channels=mid_channels,
kernel_size=1,
stride=1,
padding=0,
if_act=True,
act=act,
name=name + "_expand")
self.bottleneck_conv = ConvBNLayer(
in_channels=mid_channels,
out_channels=mid_channels,
kernel_size=kernel_size,
stride=stride,
padding=int((kernel_size - 1) // 2),
groups=mid_channels,
if_act=True,
act=act,
name=name + "_depthwise")
if self.if_se:
self.mid_se = SEModule(mid_channels, name=name + "_se")
self.linear_conv = ConvBNLayer(
in_channels=mid_channels,
out_channels=out_channels,
kernel_size=1,
stride=1,
padding=0,
if_act=False,
act=None,
name=name + "_linear")
def forward(self, inputs):
x = self.expand_conv(inputs)
x = self.bottleneck_conv(x)
if self.if_se:
x = self.mid_se(x)
x = self.linear_conv(x)
if self.if_shortcut:
x = paddle.add(inputs, x)
return x
class SEModule(nn.Layer):
def __init__(self, in_channels, reduction=4, name=""):
super(SEModule, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2D(1)
self.conv1 = nn.Conv2D(
in_channels=in_channels,
out_channels=in_channels // reduction,
kernel_size=1,
stride=1,
padding=0,
weight_attr=ParamAttr(name=name + "_1_weights"),
bias_attr=ParamAttr(name=name + "_1_offset"))
self.conv2 = nn.Conv2D(
in_channels=in_channels // reduction,
out_channels=in_channels,
kernel_size=1,
stride=1,
padding=0,
weight_attr=ParamAttr(name + "_2_weights"),
bias_attr=ParamAttr(name=name + "_2_offset"))
def forward(self, inputs):
outputs = self.avg_pool(inputs)
outputs = self.conv1(outputs)
outputs = F.relu(outputs)
outputs = self.conv2(outputs)
outputs = F.hardsigmoid(outputs, slope=0.2, offset=0.5)
return inputs * outputs
\ No newline at end of file
ppocr/modeling/backbones/table_resnet_vd.py deleted 100644 → 0
View file @ 3f3b915b
# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import paddle
from paddle import ParamAttr
import paddle.nn as nn
import paddle.nn.functional as F
__all__ = ["ResNet"]
class ConvBNLayer(nn.Layer):
def __init__(
self,
in_channels,
out_channels,
kernel_size,
stride=1,
groups=1,
is_vd_mode=False,
act=None,
name=None, ):
super(ConvBNLayer, self).__init__()
self.is_vd_mode = is_vd_mode
self._pool2d_avg = nn.AvgPool2D(
kernel_size=2, stride=2, padding=0, ceil_mode=True)
self._conv = nn.Conv2D(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=kernel_size,
stride=stride,
padding=(kernel_size - 1) // 2,
groups=groups,
weight_attr=ParamAttr(name=name + "_weights"),
bias_attr=False)
if name == "conv1":
bn_name = "bn_" + name
else:
bn_name = "bn" + name[3:]
self._batch_norm = nn.BatchNorm(
out_channels,
act=act,
param_attr=ParamAttr(name=bn_name + '_scale'),
bias_attr=ParamAttr(bn_name + '_offset'),
moving_mean_name=bn_name + '_mean',
moving_variance_name=bn_name + '_variance')
def forward(self, inputs):
if self.is_vd_mode:
inputs = self._pool2d_avg(inputs)
y = self._conv(inputs)
y = self._batch_norm(y)
return y
class BottleneckBlock(nn.Layer):
def __init__(self,
in_channels,
out_channels,
stride,
shortcut=True,
if_first=False,
name=None):
super(BottleneckBlock, self).__init__()
self.conv0 = ConvBNLayer(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=1,
act='relu',
name=name + "_branch2a")
self.conv1 = ConvBNLayer(
in_channels=out_channels,
out_channels=out_channels,
kernel_size=3,
stride=stride,
act='relu',
name=name + "_branch2b")
self.conv2 = ConvBNLayer(
in_channels=out_channels,
out_channels=out_channels * 4,
kernel_size=1,
act=None,
name=name + "_branch2c")
if not shortcut:
self.short = ConvBNLayer(
in_channels=in_channels,
out_channels=out_channels * 4,
kernel_size=1,
stride=1,
is_vd_mode=False if if_first else True,
name=name + "_branch1")
self.shortcut = shortcut
def forward(self, inputs):
y = self.conv0(inputs)
conv1 = self.conv1(y)
conv2 = self.conv2(conv1)
if self.shortcut:
short = inputs
else:
short = self.short(inputs)
y = paddle.add(x=short, y=conv2)
y = F.relu(y)
return y
class BasicBlock(nn.Layer):
def __init__(self,
in_channels,
out_channels,
stride,
shortcut=True,
if_first=False,
name=None):
super(BasicBlock, self).__init__()
self.stride = stride
self.conv0 = ConvBNLayer(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=3,
stride=stride,
act='relu',
name=name + "_branch2a")
self.conv1 = ConvBNLayer(
in_channels=out_channels,
out_channels=out_channels,
kernel_size=3,
act=None,
name=name + "_branch2b")
if not shortcut:
self.short = ConvBNLayer(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=1,
stride=1,
is_vd_mode=False if if_first else True,
name=name + "_branch1")
self.shortcut = shortcut
def forward(self, inputs):
y = self.conv0(inputs)
conv1 = self.conv1(y)
if self.shortcut:
short = inputs
else:
short = self.short(inputs)
y = paddle.add(x=short, y=conv1)
y = F.relu(y)
return y
class ResNet(nn.Layer):
def __init__(self, in_channels=3, layers=50, **kwargs):
super(ResNet, self).__init__()
self.layers = layers
supported_layers = [18, 34, 50, 101, 152, 200]
assert layers in supported_layers, \
"supported layers are {} but input layer is {}".format(
supported_layers, layers)
if layers == 18:
depth = [2, 2, 2, 2]
elif layers == 34 or layers == 50:
depth = [3, 4, 6, 3]
elif layers == 101:
depth = [3, 4, 23, 3]
elif layers == 152:
depth = [3, 8, 36, 3]
elif layers == 200:
depth = [3, 12, 48, 3]
num_channels = [64, 256, 512,
1024] if layers >= 50 else [64, 64, 128, 256]
num_filters = [64, 128, 256, 512]
self.conv1_1 = ConvBNLayer(
in_channels=in_channels,
out_channels=32,
kernel_size=3,
stride=2,
act='relu',
name="conv1_1")
self.conv1_2 = ConvBNLayer(
in_channels=32,
out_channels=32,
kernel_size=3,
stride=1,
act='relu',
name="conv1_2")
self.conv1_3 = ConvBNLayer(
in_channels=32,
out_channels=64,
kernel_size=3,
stride=1,
act='relu',
name="conv1_3")
self.pool2d_max = nn.MaxPool2D(kernel_size=3, stride=2, padding=1)
self.stages = []
self.out_channels = []
if layers >= 50:
for block in range(len(depth)):
block_list = []
shortcut = False
for i in range(depth[block]):
if layers in [101, 152] and block == 2:
if i == 0:
conv_name = "res" + str(block + 2) + "a"
else:
conv_name = "res" + str(block + 2) + "b" + str(i)
else:
conv_name = "res" + str(block + 2) + chr(97 + i)
bottleneck_block = self.add_sublayer(
'bb_%d_%d' % (block, i),
BottleneckBlock(
in_channels=num_channels[block]
if i == 0 else num_filters[block] * 4,
out_channels=num_filters[block],
stride=2 if i == 0 and block != 0 else 1,
shortcut=shortcut,
if_first=block == i == 0,
name=conv_name))
shortcut = True
block_list.append(bottleneck_block)
self.out_channels.append(num_filters[block] * 4)
self.stages.append(nn.Sequential(*block_list))
else:
for block in range(len(depth)):
block_list = []
shortcut = False
for i in range(depth[block]):
conv_name = "res" + str(block + 2) + chr(97 + i)
basic_block = self.add_sublayer(
'bb_%d_%d' % (block, i),
BasicBlock(
in_channels=num_channels[block]
if i == 0 else num_filters[block],
out_channels=num_filters[block],
stride=2 if i == 0 and block != 0 else 1,
shortcut=shortcut,
if_first=block == i == 0,
name=conv_name))
shortcut = True
block_list.append(basic_block)
self.out_channels.append(num_filters[block])
self.stages.append(nn.Sequential(*block_list))
def forward(self, inputs):
y = self.conv1_1(inputs)
y = self.conv1_2(y)
y = self.conv1_3(y)
y = self.pool2d_max(y)
out = []
for block in self.stages:
y = block(y)
out.append(y)
return out
ppocr/modeling/transforms/tps_spatial_transformer.py
View file @ 1d7c9f81
...@@ -53,7 +53,7 @@ def compute_partial_repr(input_points, control_points): ...@@ -53,7 +53,7 @@ def compute_partial_repr(input_points, control_points):
1] 1]
repr_matrix = 0.5 * pairwise_dist * paddle.log(pairwise_dist) repr_matrix = 0.5 * pairwise_dist * paddle.log(pairwise_dist)
# fix numerical error for 0 * log(0), substitute all nan with 0 # fix numerical error for 0 * log(0), substitute all nan with 0
mask = repr_matrix != repr_matrix mask = np.array(repr_matrix != repr_matrix)
repr_matrix[mask] = 0 repr_matrix[mask] = 0
return repr_matrix return repr_matrix
......
ppocr/postprocess/east_postprocess.py
View file @ 1d7c9f81
...@@ -20,7 +20,6 @@ import numpy as np ...@@ -20,7 +20,6 @@ import numpy as np
from .locality_aware_nms import nms_locality from .locality_aware_nms import nms_locality
import cv2 import cv2
import paddle import paddle
import lanms
import os import os
import sys import sys
...@@ -61,6 +60,7 @@ class EASTPostProcess(object): ...@@ -61,6 +60,7 @@ class EASTPostProcess(object):
""" """
restore text boxes from score map and geo map restore text boxes from score map and geo map
""" """
score_map = score_map[0] score_map = score_map[0]
geo_map = np.swapaxes(geo_map, 1, 0) geo_map = np.swapaxes(geo_map, 1, 0)
geo_map = np.swapaxes(geo_map, 1, 2) geo_map = np.swapaxes(geo_map, 1, 2)
...@@ -76,8 +76,15 @@ class EASTPostProcess(object): ...@@ -76,8 +76,15 @@ class EASTPostProcess(object):
boxes = np.zeros((text_box_restored.shape[0], 9), dtype=np.float32) boxes = np.zeros((text_box_restored.shape[0], 9), dtype=np.float32)
boxes[:, :8] = text_box_restored.reshape((-1, 8)) boxes[:, :8] = text_box_restored.reshape((-1, 8))
boxes[:, 8] = score_map[xy_text[:, 0], xy_text[:, 1]] boxes[:, 8] = score_map[xy_text[:, 0], xy_text[:, 1]]
boxes = lanms.merge_quadrangle_n9(boxes, nms_thresh)
# boxes = nms_locality(boxes.astype(np.float64), nms_thresh) try:
import lanms
boxes = lanms.merge_quadrangle_n9(boxes, nms_thresh)
except:
print(
'you should install lanms by pip3 install lanms-nova to speed up nms_locality'
)
boxes = nms_locality(boxes.astype(np.float64), nms_thresh)
if boxes.shape[0] == 0: if boxes.shape[0] == 0:
return [] return []
# Here we filter some low score boxes by the average score map, # Here we filter some low score boxes by the average score map,
......
ppocr/utils/save_load.py
View file @ 1d7c9f81
...@@ -67,6 +67,7 @@ def load_model(config, model, optimizer=None): ...@@ -67,6 +67,7 @@ def load_model(config, model, optimizer=None):
if key not in params: if key not in params:
logger.warning("{} not in loaded params {} !".format( logger.warning("{} not in loaded params {} !".format(
key, params.keys())) key, params.keys()))
continue
pre_value = params[key] pre_value = params[key]
if list(value.shape) == list(pre_value.shape): if list(value.shape) == list(pre_value.shape):
new_state_dict[key] = pre_value new_state_dict[key] = pre_value
...@@ -76,9 +77,14 @@ def load_model(config, model, optimizer=None): ...@@ -76,9 +77,14 @@ def load_model(config, model, optimizer=None):
format(key, value.shape, pre_value.shape)) format(key, value.shape, pre_value.shape))
model.set_state_dict(new_state_dict) model.set_state_dict(new_state_dict)
optim_dict = paddle.load(checkpoints + '.pdopt')
if optimizer is not None: if optimizer is not None:
optimizer.set_state_dict(optim_dict) if os.path.exists(checkpoints + '.pdopt'):
optim_dict = paddle.load(checkpoints + '.pdopt')
optimizer.set_state_dict(optim_dict)
else:
logger.warning(
"{}.pdopt is not exists, params of optimizer is not loaded".
format(checkpoints))
if os.path.exists(checkpoints + '.states'): if os.path.exists(checkpoints + '.states'):
with open(checkpoints + '.states', 'rb') as f: with open(checkpoints + '.states', 'rb') as f:
......
requirements.txt
View file @ 1d7c9f81
...@@ -12,5 +12,4 @@ cython ...@@ -12,5 +12,4 @@ cython
lxml lxml
premailer premailer
openpyxl openpyxl
fasttext==0.9.1 fasttext==0.9.1
lanms-nova \ No newline at end of file
\ No newline at end of file
test_tipc/configs/ch_PP-OCRv2/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt
View file @ 1d7c9f81
===========================ch_ppocr_mobile_v2.0=========================== ===========================ch_PP-OCRv2===========================
model_name:ch_PP-OCRv2 model_name:ch_PP-OCRv2
python:python3.7 python:python3.7
null:null
null:null
infer_model:./inference/ch_PP-OCRv2_det_infer/ infer_model:./inference/ch_PP-OCRv2_det_infer/
infer_export:null infer_export:null
infer_quant:False infer_quant:False
...@@ -12,7 +10,7 @@ inference:tools/infer/predict_system.py ...@@ -12,7 +10,7 @@ inference:tools/infer/predict_system.py
--cpu_threads:1|6 --cpu_threads:1|6
--rec_batch_num:1 --rec_batch_num:1
--use_tensorrt:False|True --use_tensorrt:False|True
--precision:fp32|fp16 --precision:fp32
--det_model_dir: --det_model_dir:
--image_dir:./inference/ch_det_data_50/all-sum-510/ --image_dir:./inference/ch_det_data_50/all-sum-510/
--rec_model_dir:./inference/ch_PP-OCRv2_rec_infer/ --rec_model_dir:./inference/ch_PP-OCRv2_rec_infer/
......
test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt
View file @ 1d7c9f81
===========================ch_ppocr_mobile_v2.0=========================== ===========================ch_ppocr_mobile_v2.0===========================
model_name:ch_ppocr_mobile_v2.0 model_name:ch_ppocr_mobile_v2.0
python:python3.7 python:python3.7
null:null
null:null
infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/
infer_export:null infer_export:null
infer_quant:False infer_quant:False
...@@ -12,10 +10,10 @@ inference:tools/infer/predict_system.py ...@@ -12,10 +10,10 @@ inference:tools/infer/predict_system.py
--cpu_threads:1|6 --cpu_threads:1|6
--rec_batch_num:1 --rec_batch_num:1
--use_tensorrt:False|True --use_tensorrt:False|True
--precision:fp32|fp16 --precision:fp32
--det_model_dir: --det_model_dir:
--image_dir:./inference/ch_det_data_50/all-sum-510/ --image_dir:./inference/ch_det_data_50/all-sum-510/
null:null
--benchmark:True
--rec_model_dir:./inference/ch_ppocr_mobile_v2.0_rec_infer/ --rec_model_dir:./inference/ch_ppocr_mobile_v2.0_rec_infer/
--benchmark:True
null:null
null:null null:null
test_tipc/configs/ch_ppocr_mobile_v2.0_rec/train_infer_python.txt 0 → 100644
View file @ 1d7c9f81
===========================train_params===========================
model_name:ch_ppocr_mobile_v2.0_rec
python:python3.7
gpu_list:0|0,1
Global.use_gpu:True|True
Global.auto_cast:null
Global.epoch_num:lite_train_infer=2|whole_train_infer=300
Global.save_model_dir:./output/
Train.loader.batch_size_per_card:lite_train_infer=128|whole_train_infer=128
Global.pretrained_model:null
train_model_name:latest
train_infer_img_dir:./inference/rec_inference
null:null
##
trainer:norm_train
norm_train:tools/train.py -c configs/rec/rec_icdar15_train.yml -o
pact_train:null
fpgm_train:null
distill_train:null
null:null
null:null
##
===========================eval_params===========================
eval:tools/eval.py -c configs/rec/rec_icdar15_train.yml -o
null:null
##
===========================infer_params===========================
Global.save_inference_dir:./output/
Global.pretrained_model:
norm_export:tools/export_model.py -c configs/rec/rec_icdar15_train.yml -o
quant_export:null
fpgm_export:null
distill_export:null
export1:null
export2:null
##
infer_model:null
infer_export:tools/export_model.py -c configs/rec/rec_icdar15_train.yml -o
infer_quant:False
inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ic15_dict.txt --rec_image_shape="3,32,100" --rec_algorithm="RARE"
--use_gpu:True|False
--enable_mkldnn:True|False
--cpu_threads:1|6
--rec_batch_num:1|6
--use_tensorrt:True|False
--precision:fp32|fp16|int8
--rec_model_dir:
--image_dir:./inference/rec_inference
--save_log_path:./test/output/
--benchmark:True
null:null
test_tipc/configs/ch_ppocr_mobile_v2.0_rec_FPGM/rec_chinese_lite_train_v2.0.yml 0 → 100644
View file @ 1d7c9f81
Global:
use_gpu: true
epoch_num: 500
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/rec_chinese_lite_v2.0
save_epoch_step: 3
# evaluation is run every 5000 iterations after the 4000th iteration
eval_batch_step: [0, 2000]
cal_metric_during_train: True
pretrained_model:
checkpoints:
save_inference_dir:
use_visualdl: False
infer_img: doc/imgs_words/ch/word_1.jpg
# for data or label process
character_dict_path: ppocr/utils/ppocr_keys_v1.txt
max_text_length: 25
infer_mode: False
use_space_char: True
save_res_path: ./output/rec/predicts_chinese_lite_v2.0.txt
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.001
regularizer:
name: 'L2'
factor: 0.00001
Architecture:
model_type: rec
algorithm: CRNN
Transform:
Backbone:
name: MobileNetV3
scale: 0.5
model_name: small
small_stride: [1, 2, 2, 2]
disable_se: True
Neck:
name: SequenceEncoder
encoder_type: rnn
hidden_size: 48
Head:
name: CTCHead
fc_decay: 0.00001
Loss:
name: CTCLoss
PostProcess:
name: CTCLabelDecode
Metric:
name: RecMetric
main_indicator: acc
Train:
dataset:
name: SimpleDataSet
data_dir: train_data/ic15_data
label_file_list: ["train_data/ic15_data/rec_gt_train.txt"]
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- RecAug:
- CTCLabelEncode: # Class handling label
- RecResizeImg:
image_shape: [3, 32, 320]
- KeepKeys:
keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
loader:
shuffle: True
batch_size_per_card: 256
drop_last: True
num_workers: 8
Eval:
dataset:
name: SimpleDataSet
data_dir: train_data/ic15_data
label_file_list: ["train_data/ic15_data/rec_gt_test.txt"]
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- CTCLabelEncode: # Class handling label
- RecResizeImg:
image_shape: [3, 32, 320]
- KeepKeys:
keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
loader:
shuffle: False
drop_last: False
batch_size_per_card: 256
num_workers: 8
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