KeyError: 'batch_input_shape'

[Gabriel-MarcoSilva]

independent of the backend when running:

config = hls4ml.utils.config_from_keras_model(model, granularity='model', backend='Vitis')

It keeps giving this error here:

/usr/local/lib/python3.11/dist-packages/hls4ml/converters/keras/core.py in parse_input_layer(keras_layer, input_names, input_shapes, data_reader)
10 layer = parse_default_keras_layer(keras_layer, input_names)
11
---> 12 layer['input_shape'] = keras_layer['config']['batch_input_shape'][1:]
13
14 dtype = keras_layer['config']['dtype']

KeyError: 'batch_input_shape'

for current python in case but it works on python 3.10

[vloncar]

This is likely due to the version of Keras, but I can't tell since you didn't share that information. Can you submit a bug report instead of opening a discussion thread.