Tivadar Danka

Judging from the `ValueError`, I think the problem is that the image to be saved is constant in intensity, thus it cannot convert it to uint8. I'll investigate this problem...

Hi! Although I don't have your data with me, I am pretty confident that you didn't specify `--out_channels` correctly. From the exception, it follows that the cross-entropy loss function expects...

Unfortunately, I cannot tell you the exact reason for this because I don't have enough information, but seems like your model is overfitting. Do you use the same augmentation during...

@alejandrodelac Hi! I am going to be honest, the last time I trained a U-Net on this dataset was about three years ago, so I don't exactly remember what steps...

Hi! Can you post the full stack trace? I cannot really say anything with this information only.

Hi! I am not sure what you mean by convenient to provide your training results. If I understand it correctly, training results are often reported, but validation and test losses...

According to [comments in the Keras source code](https://github.com/keras-team/keras/blob/master/keras/wrappers/scikit_learn.py#L191-L196), the scikit-learn wrapper is deprecated and the [scikeras](https://github.com/adriangb/scikeras) external lib should be used. (As I was not following Keras development, this surprised...

Hi! I haven't planned this feature, but this would be useful indeed. Meanwhile, if you need this feature urgently, the individual query strategies can be easily modified to achieve this....

Sorry for the delayed answer, I was extremely busy in the past few days. I am not sure what would be the good design for this feature. One alternative would...

Hi! What is the bug exactly? Can you copy and paste some code snippets and error logs here? Thanks!