提交 6b208bcd 编写于 作者: T tink2123

fix typo

上级 0cc083f6
......@@ -36,7 +36,8 @@ PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库,助力
#### 2.inference模型下载
*windows 环境下没有如果没有安装wget,下载模型时可将链接复制到浏览器中下载,并解压放置在相应目录下*
*windows 环境下如果没有安装wget,下载模型时可将链接复制到浏览器中下载,并解压放置在相应目录下*
#### (1)超轻量级中文OCR模型下载
```
......
TrainReader:
reader_function: ppocr.data.rec.dataset_traversal,LMDBReader
num_workers: 8
lmdb_sets_dir: ./train_data/data_lmdb_release/evaluation/
lmdb_sets_dir: ./train_data/data_lmdb_release/training/
EvalReader:
reader_function: ppocr.data.rec.dataset_traversal,LMDBReader
......
Global:
algorithm: CRNN
use_gpu: false
use_gpu: true
epoch_num: 1000
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/rec_CRNN
save_epoch_step: 300
eval_batch_step: 500
train_batch_size_per_card: 2
test_batch_size_per_card: 2
train_batch_size_per_card: 256
test_batch_size_per_card: 256
image_shape: [3, 32, 100]
max_text_length: 25
character_type: en
loss_type: ctc
reader_yml: ./configs/rec/rec_icdar15_reader.yml
pretrain_weights:
pretrain_weights: ./pretrain_models/rec_mv3_none_bilstm_ctc/best_accuracy
checkpoints:
save_inference_dir:
infer_img:
......
Global:
algorithm: RARE
use_gpu: false
use_gpu: true
epoch_num: 72
log_smooth_window: 20
print_batch_step: 10
save_model_dir: output/rec_RARE
save_epoch_step: 3
eval_batch_step: 2000
train_batch_size_per_card: 2
test_batch_size_per_card: 2
train_batch_size_per_card: 256
test_batch_size_per_card: 256
image_shape: [3, 32, 100]
max_text_length: 25
character_type: en
......
......@@ -46,6 +46,9 @@ wget -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/Res
```
**启动训练**
*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*
```
python3 tools/train.py -c configs/det/det_mv3_db.yml
```
......
......@@ -113,6 +113,8 @@ tar -xf rec_mv3_none_bilstm_ctc.tar && rm -rf rec_mv3_none_bilstm_ctc.tar
开始训练:
*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*
```
# 设置PYTHONPATH路径
export PYTHONPATH=$PYTHONPATH:.
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册