sushreebarsa comments

Results 267 comments of


                                            sushreebarsa

The loss displayed by the network fit is inconsistent with the loss calculated from the input data after the model is saved.

@ewwll Could you please confirm if it is a duplicate of [this](https://github.com/keras-team/keras/issues/16836) ticket? Thank you!

Cannot Build Intermediate Model to Nested Layers

@jvishnuvardhan I was able to reproduce this issue on colab using TF v2.8.0 and tf-nightly ,please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/e381933d02db9422fa2aef63edaa6b19/16123.ipynb#scrollTo=toKLGrJaCs4_).Thanks!

Grappler Errors with ConvLSTM2D

@acfrock I tried to run the example script for replicating the issue and didn’t face the error reported. Could you find the gist [here](https://colab.research.google.com/gist/sushreebarsa/917d8ee540e79e34a57e091360bd45a3/conv_lstm.ipynb#scrollTo=N4p596xMrKhY) and let me know if I...

Mask that had been removed from a tensor re-appears when building a new model with placeholders

@gadagashwini I was able to replicate this issue on colab using TF v2.8.0 and tf-nightly,please find the gist[ here](https://colab.research.google.com/gist/sushreebarsa/9cf59b97f31758483496dc727ec1e51f/16441.ipynb).Thanks!

Unpredictable behaviour of tracked variables in subclassing layers and models

@gadagashwini I was able to replicate the issue on colab, please find the gist[ here](https://colab.research.google.com/gist/sushreebarsa/0f943c1c82ee5f907eb3c59a7b9d79f7/possible_keras_bug_subclassing.ipynb). Thank you!

kernel dimensions in Depthwiseconv1d()

@dario-passos Could you please provide more information on the issue and share the documentation site that you are referring? Thank you!

Using output names instead of node names for loss, exporting and serving

@gadagashwini I was able to replicate the issue, please find the gist [here ](https://colab.research.google.com/gist/sushreebarsa/bde7cd6c2455631273a6acf620bf8934/16758.ipynb). Thank you!

`model.validate` resets training metrics

@diggerk In order to expedite the trouble-shooting process, please provide a complete code snippet to reproduce the issue reported here.I am facing a different [error](https://colab.research.google.com/gist/sushreebarsa/8c6b6be7afc4497fa500f6c1bc5381da/16526.ipynb#scrollTo=nlTbDqkSNcBZ) while reproducing this issue. Thank...

`model.validate` resets training metrics

@diggerk Reopening this issue as per the above [comment](https://github.com/keras-team/keras/issues/16526#issuecomment-1152724930). Thank you!

`model.validate` resets training metrics

@gadagashwini I was able to replicate the issue on colab, please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/1234385d07e794c278ab00600214df09/highres-metrics-with-mnist.ipynb). Thank you!