Jithendra Paruchuri

Apple silicon uses pip install tensorflow-macos , the regular tensorflow requirement causing issue

~/june/git/onnx2keras/onnx2keras/converter.py in onnx_to_keras(onnx_model, input_names, input_shapes, name_policy, verbose, change_ordering) 275 276 keras.backend.set_image_data_format('channels_last') --> 277 model_tf_ordering = keras.models.Model.from_config(conf) 278 279 for dst_layer, src_layer, conf in zip(model_tf_ordering.layers, model.layers, conf['layers']): ~/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py in from_config(cls, config,...

This is not an issue but a feature request :)

Conversion using tensorflow.python.keras.layers.normalization.BatchNormalization not v2

1

@nerox8664 In TF 2.x, its suppose to use tensorflow.python.keras.layers.normalization_v2.BatchNormalization but looks like this conversion is using older tensorflow.python.keras.layers.normalization.BatchNormalization. Tf version: 2.1.0