search

nabsabraham / focal-tversky-unet

This repo contains the code for our paper "A novel focal Tversky loss function and improved Attention U-Net for lesion segmentation" accepted at IEEE ISBI 2019.
357 stars 72 forks source link

Version issue #14

Closed DanielLee343 closed 4 years ago

DanielLee343 commented 4 years ago

Hi, hope you see this. Which version of TensorFlow and python did you use to train the model, 'cause my version fails to execute the following code: precision, recall, thresholds = precision_recall_curve(y_true, y_preds) in training isic dataset. Guess it might be some version issues. Thanks!

nabsabraham commented 4 years ago

Hi Daniel! I believe it was tf 1.0.1

Halleyawoo commented 4 years ago

Hi, hope you see this. Which version of TensorFlow and python did you use to train the model, 'cause my version fails to execute the following code: precision, recall, thresholds = precision_recall_curve(y_true, y_preds) in training isic dataset. Guess it might be some version issues. Thanks!

Hello, have you solved this problem？

DanielLee343 commented 4 years ago

Yes. My configuration is tensorflow-gpu = 1.13.1, keras = 2.3.1, cuda = 10.1 and it works fine then.

Halleyawoo commented 4 years ago

Yes. My configuration is tensorflow-gpu = 1.13.1, keras = 2.3.1, cuda = 10.1 and it works fine then.

博主你好，我的版本是tensorflow-gpu = 1.0.1, keras = 2.0.2, cuda = 8.0，不知道是不是要全部和你改成一样的，方便的话可以展示一下最后的结果图吗（就是代码最后#plot precision-recall 这里的）？

Halleyawoo commented 4 years ago

Yes. My configuration is tensorflow-gpu = 1.13.1, keras = 2.3.1, cuda = 10.1 and it works fine then.

谢谢你的回答

DanielLee343 commented 4 years ago

最好改成一样的吧，改一下很容易，不然到时候出bug都不知道是哪里出错

Halleyawoo commented 4 years ago

最好改成一样的吧，改一下很容易，不然到时候出bug都不知道是哪里出错

好的，多谢回答。

Halleyawoo commented 4 years ago

大佬，请问那个分割结果的可视化你做了吗，怎么实现呢