search

SciSharp / TensorFlow.NET

.NET Standard bindings for Google's TensorFlow for developing, training and deploying Machine Learning models in C# and F#.
https://scisharp.github.io/tensorflow-net-docs
Apache License 2.0
3.23k stars 517 forks source link

[BUG Report]: Repeated progress bar in Model.fit() #1121

Closed Tvvgge closed 1 year ago

Tvvgge commented 1 year ago

Description

In Tensorflow.Keras 0.10.5 the progress bar look like this: Epoch: 001/100 0009/0009 [=========================>...] - 2ms/step - loss: 20.894537 - val_loss: 31.065659 Epoch: 002/100 0009/0009 [=========================>...] - 4ms/step - loss: 19.879198 - val_loss: 30.460979 Epoch: 003/100 0009/0009 [=========================>...] - 2ms/step - loss: 18.928097 - val_loss: 29.738377

When upgraded to 0.11.0 the progress bar becomes: Epoch: 001/100 Epoch: 001/000======================>...] - 6ms/step - loss: 20.9055106 0009/0009 [=========================>...] - 6ms/step - loss: 20.905510 - val_loss: 30.871361 Epoch: 002/100 Epoch: 001/000======================>...] - 2ms/step - loss: 19.939381 0009/0009 [=========================>...] - 2ms/step - loss: 19.939381 - val_loss: 30.301033 Epoch: 003/100 Epoch: 001/000======================>...] - 2ms/step - loss: 18.945526 0009/0009 [=========================>...] - 2ms/step - loss: 18.945526 - val_loss: 29.720556

Reproduction Steps

Upgrade the Tensorflow.Keras to version 0.11.0

Known Workarounds

No response

Configuration and Other Information

No response

Wanglongzhi2001 commented 1 year ago

Thanks, I will fix it later.

Wanglongzhi2001 commented 1 year ago

Hello, I have fixed this bug. ^_^