Comments (4)
2022-07-28 11:25:37.843 | INFO | utils.train:start:108 - [2022-07-28-11_25_37] Epoch: 36 Step: 31800 LastLoss: 0.0010917907347902656 AvgLoss: 0.0009280425880569965 Lr: 0.007385691026454037
2022-07-28 11:25:39.978 | INFO | utils.train:start:108 - [2022-07-28-11_25_39] Epoch: 36 Step: 31900 LastLoss: 0.0008689984097145498 AvgLoss: 0.0008995186287211254 Lr: 0.007385691026454037
2022-07-28 11:25:42.212 | INFO | utils.train:start:137 - [2022-07-28-11_25_42] Epoch: 36 Step: 32000 LastLoss: 0.0010542419040575624 AvgLoss: 0.0008768078277353197 Lr: 0.007385691026454037 Acc: 0.90625
2022-07-28 11:25:44.384 | INFO | utils.train:start:108 - [2022-07-28-11_25_44] Epoch: 36 Step: 32100 LastLoss: 0.0006970251561142504 AvgLoss: 0.0008913403417682276 Lr: 0.0072379772059249555
2022-07-28 11:25:46.542 | INFO | utils.train:start:108 - [2022-07-28-11_25_46] Epoch: 36 Step: 32200 LastLoss: 0.0012068169889971614 AvgLoss: 0.0008795152555103414 Lr: 0.0072379772059249555
2022-07-28 11:25:48.683 | INFO | utils.train:start:108 - [2022-07-28-11_25_48] Epoch: 36 Step: 32300 LastLoss: 0.0005893478519283235 AvgLoss: 0.0008610051040886901 Lr: 0.0072379772059249555
2022-07-28 11:25:50.824 | INFO | utils.train:start:108 - [2022-07-28-11_25_50] Epoch: 36 Step: 32400 LastLoss: 0.0008560416754335165 AvgLoss: 0.0008518385927891359 Lr: 0.0072379772059249555
2022-07-28 11:25:52.967 | INFO | utils.train:start:108 - [2022-07-28-11_25_52] Epoch: 36 Step: 32500 LastLoss: 0.0007048668339848518 AvgLoss: 0.0008719124580966309 Lr: 0.0072379772059249555
2022-07-28 11:25:55.062 | INFO | utils.train:start:108 - [2022-07-28-11_25_55] Epoch: 37 Step: 32600 LastLoss: 0.0006410101777873933 AvgLoss: 0.0008731031231582164 Lr: 0.0072379772059249555
2022-07-28 11:25:57.213 | INFO | utils.train:start:108 - [2022-07-28-11_25_57] Epoch: 37 Step: 32700 LastLoss: 0.0009320578537881374 AvgLoss: 0.0008294750811182894 Lr: 0.0072379772059249555
2022-07-28 11:25:59.322 | INFO | utils.train:start:108 - [2022-07-28-11_25_59] Epoch: 37 Step: 32800 LastLoss: 0.0009776148945093155 AvgLoss: 0.000850687074707821 Lr: 0.0072379772059249555
2022-07-28 11:26:01.465 | INFO | utils.train:start:108 - [2022-07-28-11_26_01] Epoch: 37 Step: 32900 LastLoss: 0.0013811136595904827 AvgLoss: 0.0008380734102684073 Lr: 0.0072379772059249555
2022-07-28 11:26:03.674 | INFO | utils.train:start:137 - [2022-07-28-11_26_03] Epoch: 37 Step: 33000 LastLoss: 0.000778629386331886 AvgLoss: 0.000821824642480351 Lr: 0.0072379772059249555 Acc: 0.90625
2022-07-28 11:26:05.869 | INFO | utils.train:start:108 - [2022-07-28-11_26_05] Epoch: 37 Step: 33100 LastLoss: 0.000949505774769932 AvgLoss: 0.0008630690455902368 Lr: 0.0072379772059249555
2022-07-28 11:26:08.024 | INFO | utils.train:start:108 - [2022-07-28-11_26_08] Epoch: 37 Step: 33200 LastLoss: 0.0009300833917222917 AvgLoss: 0.0008534722673357464 Lr: 0.0072379772059249555
2022-07-28 11:26:10.182 | INFO | utils.train:start:108 - [2022-07-28-11_26_10] Epoch: 37 Step: 33300 LastLoss: 0.0007823986816219985 AvgLoss: 0.0008065305658965372 Lr: 0.0072379772059249555
2022-07-28 11:26:12.454 | INFO | utils.train:start:108 - [2022-07-28-11_26_12] Epoch: 37 Step: 33400 LastLoss: 0.0007098791538737714 AvgLoss: 0.0008171254771878011 Lr: 0.0072379772059249555
2022-07-28 11:26:14.961 | INFO | utils.train:start:108 - [2022-07-28-11_26_14] Epoch: 38 Step: 33500 LastLoss: 0.001108370954170823 AvgLoss: 0.0008040564699331299 Lr: 0.0072379772059249555
2022-07-28 11:26:17.309 | INFO | utils.train:start:108 - [2022-07-28-11_26_17] Epoch: 38 Step: 33600 LastLoss: 0.0012630750425159931 AvgLoss: 0.0007878648146288469 Lr: 0.0072379772059249555
2022-07-28 11:26:20.348 | INFO | utils.train:start:108 - [2022-07-28-11_26_20] Epoch: 38 Step: 33700 LastLoss: 0.0008831340819597244 AvgLoss: 0.0007882761370274238 Lr: 0.0072379772059249555
2022-07-28 11:26:24.197 | INFO | utils.train:start:108 - [2022-07-28-11_26_24] Epoch: 38 Step: 33800 LastLoss: 0.0005721147754229605 AvgLoss: 0.0008209841320058331 Lr: 0.0072379772059249555
2022-07-28 11:26:28.109 | INFO | utils.train:start:108 - [2022-07-28-11_26_28] Epoch: 38 Step: 33900 LastLoss: 0.0007059407653287053 AvgLoss: 0.0008052434003911912 Lr: 0.0072379772059249555
2022-07-28 11:26:32.127 | INFO | utils.train:start:137 - [2022-07-28-11_26_32] Epoch: 38 Step: 34000 LastLoss: 0.0009513929835520685 AvgLoss: 0.0008047035793424584 Lr: 0.0072379772059249555 Acc: 1.0
2022-07-28 11:26:32.142 | INFO | utils.train:start:143 -
Training Finished!Exporting Model...
Traceback (most recent call last):
File "app.py", line 33, in
fire.Fire(App)
File "C:\Users\yxn\AppData\Local\Programs\Python\Python38\lib\site-packages\fire\core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "C:\Users\yxn\AppData\Local\Programs\Python\Python38\lib\site-packages\fire\core.py", line 466, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "C:\Users\yxn\AppData\Local\Programs\Python\Python38\lib\site-packages\fire\core.py", line 681, in CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "app.py", line 28, in train
trainer.start()
File "D:\works\dddd_trainer\utils\train.py", line 152, in start
self.net.export_onnx(self.net, dummy_input,
File "D:\works\dddd_trainer\nets_init.py", line 216, in export_onnx
torch.onnx.export(net, dummy_input, graph_path, export_params=True, verbose=False,
TypeError: export() got an unexpected keyword argument '_retain_param_name'
训练完遇到了同样的问题,数据集我用的captcha库生成的4位验证码,大概有2.8w张,前后运行了两次,均提示上述错误,求大佬指导下。。。
我的显卡RTX A2000 12GB,安装的cuba是11.7.57版本,cuddn正常安装,pytorch是1.12.0+cu116版本,python是3.8.10版本
继续补充:把_retain_param_name=false删掉后就不报错了,能够生成模型了,但是会有如下警告:
C:\Users\xxx\AppData\Local\Programs\Python\Python38\lib\site-packages\torch\onnx\symbolic_opset9.py:3220: UserWarning: Exporting a model to ONNX with a batch_size other than 1, with a variable length with LSTM can cause an error when running the ONNX model with a different batch size. Make sure to save the model with a batch size of 1, or define the initial states (h0/c0) as inputs of the model.
warnings.warn(
WARNING: The shape inference of prim::Constant type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of prim::Constant type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of prim::Constant type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
继续补充2:用上述训练成功的模型,结合ddddocr的sdk,能够顺利识别验证码,撒花~
from dddd_trainer.
Training Finished!Exporting Model...
Traceback (most recent call last):
File "app.py", line 33, in
fire.Fire(App)
File "/usr/local/lib/python3.8/dist-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/usr/local/lib/python3.8/dist-packages/fire/core.py", line 466, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/usr/local/lib/python3.8/dist-packages/fire/core.py", line 681, in _CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "app.py", line 28, in train
trainer.start()
File "/opt/dddd_trainer/utils/train.py", line 152, in start
self.net.export_onnx(self.net, dummy_input,
File "/opt/dddd_trainer/nets/init.py", line 216, in export_onnx
torch.onnx.export(net, dummy_input, graph_path, export_params=True, verbose=False,
TypeError: export() got an unexpected keyword argument '_retain_param_name'
Ubuntu 训练完同样
from dddd_trainer.
Training Finished!Exporting Model...
Traceback (most recent call last):
File "E:\dddd_trainer-main\app.py", line 33, in
fire.Fire(App)
File "D:\Program Files\Python310\lib\site-packages\fire\core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "D:\Program Files\Python310\lib\site-packages\fire\core.py", line 466, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "D:\Program Files\Python310\lib\site-packages\fire\core.py", line 681, in CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "E:\dddd_trainer-main\app.py", line 28, in train
trainer.start()
File "E:\dddd_trainer-main\utils\train.py", line 152, in start
self.net.export_onnx(self.net, dummy_input,
File "E:\dddd_trainer-main\nets_init.py", line 216, in export_onnx
torch.onnx.export(net, dummy_input, graph_path, export_params=True, verbose=False,
TypeError: export() got an unexpected keyword argument '_retain_param_name'
PS E:\dddd_trainer-main>
window 10 3090 cuda11.6 同样训练完毕导出的时候报异常
from dddd_trainer.
直接删掉 _retain_param_name
from dddd_trainer.
Related Issues (20)
- 有可以提供参考的训练时长吗 HOT 3
- 清晰空心数字无法识别 HOT 2
- 是否支持滑块验证码学习? HOT 4
- 导出时报错:TypeError: export() got an unexpected keyword argument '_retain_param_name'
- CtrlC中断训练后,重新执行训练命令后报错
- 请问CPU训练只有阶段数据没有模型数据为何 HOT 2
- 能放出来同花顺客户端训练的权重吗?
- 这个报错是啥意思 HOT 2
- 因磁盘满而中断后,无法自动恢复 HOT 1
- Mac也是按这个步骤来吗,改成CPU训练就可以了嘛 HOT 6
- 电脑是2060显卡能搞吗 HOT 4
- 断点恢复训练问题
- 学习素材怎么准备 HOT 2
- 可以提供坐标点选的训练吗 HOT 1
- 将ddddocr换成其他CNN比如mobilenetv2会变成NAN
- 训练了一天英文数字验证码,正确率真的感人,是我电脑不行吗....
- 训练测试集一段时间报错
- StopIteration 問題
- 希望能出一版colab上可以运行的ipynb代码
- acc准确率始终是0
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from dddd_trainer.