训练自己数据集时没有loss的log信息打印出来 #271
Comments
|
这种情况应该是数据读取有误，可以更新一下代码，根据打印出的报错信息修改数据读取路径。或者参考 #115 手动添加报错信息。 |
|
应该不是那个问题，没有执行到那一步 |
|
是因为没有读取到数据啊，所以网络没有开始计算结果。先更新一下代码试试吧～ |
|
你们代码try,except异常记录下log吧，不然还要回头看代码 |
我们后续考虑优化下reader部分的代码，增加报错信息等。目前看上去你的问题还是数据读取的问题，建议先调试下数据读取器是否能正常加载数据。 |
|
有点看不懂这个报错： |
|
然后是这样的，我是真看不懂 |
|
你的split/train.txt 里面的数据格式是什么样的？是否符合icdar15的数据格式？ 另外，你的配置文件好像有问题，你是想训练backbone为MV3_large的DB模型？训练DB模型需要用DB的后处理和预处理，看你的reader里选用的是EAST的处理函数； |
|
问题解决了，太感谢你们了！反馈真的快！ |
an1018
pushed a commit
to an1018/PaddleOCR
that referenced
this issue
Aug 17, 2022
* [cherry-pick] Fix v0.2 docs (PaddlePaddle#258) * fix save checkpoint in quantization (PaddlePaddle#257) * fix save checkpoint in quantization * fix details * Fix softlink (PaddlePaddle#256) Co-authored-by: Liufang Sang <[email protected]> * [cherry-pick] add prune benchmark (PaddlePaddle#262) * add prune model latency (PaddlePaddle#260) & fix run command for windows (PaddlePaddle#261) * Cherry-pick update reader config (PaddlePaddle#269) * Force `cudnn` backend for depthwise convs when fp16 is enabled (PaddlePaddle#271) * fix weights path (PaddlePaddle#274) * cherry-pick refine fruit config (PaddlePaddle#289) * cherry-pick polish doc (PaddlePaddle#287) * cherry-pick fix oidv5 link (PaddlePaddle#296) cherry-pick fix oidv5 link * [cherry-pick] polish slim model zoo (PaddlePaddle#313) * polish slim README * refine pedestrian_yolov3_darknet.yml and vehicle_yolov3_darknet.yml in contrib (PaddlePaddle#324) * fix yolo configs (PaddlePaddle#332) * cherry pick polish doc (PaddlePaddle#329) * refine config (PaddlePaddle#336) * [cherry pick] refine sensitive link (PaddlePaddle#334) * refine sensitive link * add YOLOv3Loss.batch_size comments (PaddlePaddle#348) * add deploy with python * update README * support TRT and Executor * update cpp_infer for paddle latest * support fixed shape of model * polish code * supoort trt * support TRT&Executor * deploy/python : fix coding style problems * update requirements.txt * update deploy/python docs Co-authored-by: Guanghua Yu <[email protected]> Co-authored-by: Liufang Sang <[email protected]> Co-authored-by: Kaipeng Deng <[email protected]> Co-authored-by: wangguanzhong <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: littletomatodonkey <[email protected]> Co-authored-by: FlyingQianMM <[email protected]> Co-authored-by: Channingss <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The text was updated successfully, but these errors were encountered: