sushreebarsa
sushreebarsa
@irskid5 For more information on custom losses,you may refer to [this](https://keras.io/api/losses/) link. If the issue still persists for you then please post this issue on [keras-team/keras repo.](https://github.com/keras-team/keras/issues) as Keras development...
@SuryanarayanaY I was able to replicate this issue on colab, please find the [gist](https://colab.research.google.com/gist/sushreebarsa/a1d979730dd181f2ec56a21ac66ce5b1/untitled766.ipynb#scrollTo=b5aU8EL8-iBq) here. Thank you!
@mbarnes13 In order to expedite the trouble-shooting process, please provide some use cases for this feature . It will help us to analyze the issue. Thank you!
I ran the code in TF v2.5 and face error ,please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/0bf4ae93db25542263e443bad439fb94/untitled155.ipynb?authuser=1)..Thanks !
@trickiwoo Could you please let us know if the above workaround worked for you? Thank you!
@pjpratik I was able to replicate the issue, please find the [gist](https://colab.research.google.com/gist/sushreebarsa/d81e141e46491014502142931af458bc/61290.ipynb) here. Thank you!
@lcnature TensorFlow's tf.gradients function can be used to numerically approximate the gradients of these functions. TF v 1.x is not actively supported and please check the migration [doc](https://www.tensorflow.org/guide/migrate). If you...
@gfolego In TensorFlow in_top_k function doesn't have a built-in option to exclude ties. Maybe one workaround can be as follows to Filter Ties Manually by obtaining the indices of the...
@abishgj-rzt One potential approach can be Customize ChiefSessionCreator. Could you try to do this by creating a custom ChiefSessionCreator subclass and within its create_session method, implement a retry loop with...
@yaroslavvb TF v1.x is not actively supported and tf.contrib is no longer supported in TensorFlow 2.x. Here is the [migration](https://www.tensorflow.org/guide/migrate) guide. Please create a new issue if it is still...