soumik12345 / nearest-celebrity-face Goto Github PK

I tried to run the code after porting it to Tensorflow 2.0 and for that, I only made changes in the imports. But while creating the model object it throws an error for shape mismatch in feature concatenation operation in inception_block_1a. Can you please check?

Environment:
Python==3.6.8
Tensorflow==2.1.0

But I think the environment has nothing to do with shape mismatch in the model.

Full traceback of the error:

Traceback (most recent call last): File "inception_blocks_v2.py", line 270, in <module> inception_resnet_v2((3, 96, 96)) File "inception_blocks_v2.py", line 246, in inception_resnet_v2 X = inception_block_1b(X) File "inception_blocks_v2.py", line 75, in inception_block_1b inception = concatenate([X_3x3, X_5x5, X_pool, X_1x1], axis=-1) File "/home/rishab/anaconda3/envs/fr_env/lib/python3.6/site-packages/tensorflow_core/python/keras/layers/merge.py", line 716, in concatenate return Concatenate(axis=axis, **kwargs)(inputs) File "/home/rishab/anaconda3/envs/fr_env/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 748, in __call__ self._maybe_build(inputs) File "/home/rishab/anaconda3/envs/fr_env/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 2116, in _maybe_build self.build(input_shapes) File "/home/rishab/anaconda3/envs/fr_env/lib/python3.6/site-packages/tensorflow_core/python/keras/utils/tf_utils.py", line 306, in wrapper output_shape = fn(instance, input_shape) File "/home/rishab/anaconda3/envs/fr_env/lib/python3.6/site-packages/tensorflow_core/python/keras/layers/merge.py", line 401, in build raise ValueError(err_msg) ValueError: A Concatenate layer requires inputs with matching shapes except for the concat axis. Got inputs shapes: [(None, 128, 12, 12), (None, 64, 12, 12), (None, 64, 12, 12), (None, 64, 12, 12)]

Thanks in advance.
Any help is appreciated.

I use python3 with all dependencies please resolve my issue