Bita Hasheminezhad

@hkaiser, I added the link to their description in the Keras documentation which contains their numpy implementation. Do you need the actual math behind them from scholar articles, too?

To support Keras, we can also add: - [x] [one_hot](https://keras.io/backend/#one_hot) (#871) - [x] [categorical_crossentropy](https://keras.io/backend/#categorical_crossentropy), implementation in [NumPy backend](https://github.com/STEllAR-GROUP/keras/blob/master/keras/backend/numpy_backend.py#L327) (#893) - [x] [binary_crossentropy](https://keras.io/backend/#binary_crossentropy), implementation in [NumPy backend](https://github.com/STEllAR-GROUP/keras/blob/master/keras/backend/numpy_backend.py#L325) (#945) - [ ]...

Keras slice has the same functionality as [TensorFlow slice](https://www.tensorflow.org/api_docs/python/tf/slice). Having an nD input array, `start` and `size` are vectors of size n. To slice a matrix, first we find the...

Keras backend has two functions for shape. The [shape](https://keras.io/backend/#shape) returns the symbolic shape of a tensor or variable. TensorFlow returns something like `` which includes the name ('Shape_8:0`), the shape...

To enlarge images, Keras uses [resize_images](https://keras.io/backend/#resize_images). Enlargement can be along the height, width or both and is to determine with `height_factor` and `width_factor`. `data_format` tells us which convention is used...

[map_fn](https://keras.io/backend/#map_fn), [foldl](https://keras.io/backend/#foldl) and [foldr](https://keras.io/backend/#foldr) have two common arguments, `fn` (callable) and `elems`. Here is an example of how `foldl` is used: ```py import numpy as np from phylanx import Phylanx,...

Keras [switch](https://keras.io/backend/#switch) is basically an element-wise `if_conditional`. If the condition has lower dimensions in comparison to `then` and `else`, the condition will be broadcasted (it will be reshaped to (oldshape,...

@hkaiser, are changes on tests/unit/distributed/remote_add.cpp on purpose?

@scheherzade, thanks for your effort to make it compatible. It raises a TypeError, ```pytb decode_pred_tf, log_prob_pred_tf = K.ctc_decode(inputs, > input_length, greedy=True) E TypeError: 'eval_wrapper' object is not iterable tests\keras\backend\backend_test.py:1748: TypeError...

Here is an example that should produce two output tensors.