keras

keras copied to clipboard

[Refactoring] making the code more Succinct and Pythonic

2

This PR uses the `with` statement and the `.join` function to make the code more Pythonic and Succinct.

maldil

Allow ModelCheckpoint callback to save Tensorflow SavedModel

5

**Here's why we have that policy:**. Keras developers respond to issues. We want to focus on work that benefits the whole community, e.g., fixing bugs and adding features. Support only...

jordisoler

Optimizer backward compatibility fixes. 1. Add back the return value of `apply_gradients()` to keep backward compatibility. 2. Add back the name arg of `apply_gradients()`.

copybara-service[bot]

I am trying to create a multi-class classification model with Keras and it appears that when I try to fit the model, the shapes in the output are different that...

kkadu

Add f-string format and check with flynt for the whole codebase

4

Hi, In order to standardize the f-string usage of the whole codebase I have added [flynt](https://github.com/ikamensh/flynt) to the `format` and `lint` scripts. This avoids multiple ["f-string"](https://github.com/keras-team/keras/pulls?page=1&q=f-string) one line pull requests...

edumucelli

configurable `distribute_reduction_method` in Model.

30

additionally allows to reduce with `sum` to account for the metrics/losses divided by global batch size - when training using custom training step and MirroredStrategy. This addresses https://github.com/keras-team/keras/issues/15509

kretes

**System information**. TensorFlow version (you are using): 2.8.2 Are you willing to contribute it (Yes/No) : yes **Describe the feature and the current behavior/state**. Currently, when trying to fit a...

emailic

PR for solving issue #16797

5

A minimal change that solves the issue is modifying the method ```keras.layers.Layer.trainable_weights()```, in base_layer.py, by replacing it with the analogous method of keras.Model. But still I have some questions: Does...

JaimeArboleda

TimeDistributed model compatibility

6

**System information**. - Have I written custom code (as opposed to using a stock example script provided in Keras): Yes - OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows...

andreped

Feature request: optimizer support with slot variables in different dtype to the variable.

5

Currently an assumption is made that slot variables have the same dtype as the variable, which prevents experimentation. From digging around a solution is to: 1. Extend `SlotVariableReference` proto in...

georgepaw