Update README.md

README.md

@@ -23,7 +23,7 @@ CRNN（卷积循环神经网络）是一种用于光学字符识别（OCR）的
 
 ## 环境配置
 ### Docker（方法一）
-推荐使用docker方式运行， 此处提供[光源](https://www.sourcefind.cn/#/service-details)拉取docker镜像的地址与使用步骤
+推荐使用docker方式运行， 此处提供[光源](https://www.sourcefind.cn/#/service-list)拉取docker镜像的地址与使用步骤
 ```
 docker pull image.sourcefind.cn:5000/dcu/admin/base/paddlepaddle:3.0.0-py3.10-dtk24.04.3-ubuntu20.04
 
@@ -92,13 +92,14 @@ export LD_LIBRARY_PATH=/opt/dtk-24.04.3/lib:/opt/dtk-24.04.3/cuda/extras/CUPTI/l
 训练集应有如下文件结构：
 ```
 |-train_data
-  |-rec
-    |- rec_gt_train.txt
-    |- train
-        |- word_001.png
-        |- word_002.jpg
-        |- word_003.jpg
-        | ...
+    |-icdar2015
+        |-rec
+            |- rec_gt_train.txt
+            |- train
+                |- word_001.png
+                |- word_002.jpg
+                |- word_003.jpg
+                | ...
 ```
 
 ## 训练
@@ -123,7 +124,7 @@ configs/rec/rec_r34_vd_none_bilstm_ctc.yml
 ```
 ### 运行训练文件
 ```
-CUDA_VISIBLE_DEVICES=1 python3 tools/train.py -c ./PaddleOCR-main/configs/rec/rec_r34_vd_none_bilstm_ctc.yml  -o Global.pretrained_model=./PaddleOCR-main/pretrain_models/rec_mv3_none_bilstm_ctc_v2.0_train/best_accuracy
+CUDA_VISIBLE_DEVICES=1 python3 tools/train.py -c ./configs/rec/rec_r34_vd_none_bilstm_ctc.yml  -o Global.pretrained_model=./pretrain_models/rec_r34_vd_none_bilstm_ctc_v2.0_train/best_accuracy
 ```
 
 
@@ -135,7 +136,7 @@ CUDA_VISIBLE_DEVICES=1 python3 tools/train.py -c ./PaddleOCR-main/configs/rec/re
 
 ```
 # 预测使用的配置文件必须与训练一致
-python3 tools/infer_rec.py -c configs/rec/rec_r34_vd_none_bilstm_ctc.yml -o Global.pretrained_model='./PaddleOCR-main/pretrain_models/rec_r34_vd_none_bilstm_ctc_v2.0_train/best_accuracy' Global.infer_img='./PaddleOCR-main/docs/images/ppocrv4.png'
+python3 tools/infer_rec.py -c configs/rec/rec_r34_vd_none_bilstm_ctc.yml -o Global.pretrained_model='./pretrain_models/rec_r34_vd_none_bilstm_ctc_v2.0_train/best_accuracy' Global.infer_img='./docs/images/ppocrv4.png'
 ```
 
 
@@ -164,7 +165,7 @@ python3 tools/infer_rec.py -c configs/rec/rec_r34_vd_none_bilstm_ctc.yml -o Glob
 ## 预训练权重
 
 
-CRNN PaddlePaddle下载地址为：[rec_r34_vd_none_bilstm_ctc.yml](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar)
+CRNN PaddlePaddle下载地址为：[rec_r34_vd_none_bilstm_ctc](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar)
 
 
 ## 源码仓库及问题反馈