The input must have 3 channels, I need 1 channel

[MohannadEhabBarakat]

When running following code base_model = efn.EfficientNetB0(include_top=False, weights='imagenet', input_shape=(256,256,1)) I get Traceback (most recent call last): File "models/discriminator.py", line 40, in model = discriminator((256,256,1)) File "models/discriminator.py", line 19, in discriminator base_model.input = layers.Input(input_shape) File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/keras/engine/network.py", line 323, in setattr super(Network, self).setattr(name, value) File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/keras/engine/base_layer.py", line 1245, in setattr super(Layer, self).setattr(name, value) AttributeError: can't set attribute (tf1) mbarakat@ml-server2:~/gan$ python models/discriminator.py Using TensorFlow backend. Traceback (most recent call last): File "models/discriminator.py", line 40, in model = discriminator((256,256,1)) File "models/discriminator.py", line 11, in discriminator base_model = efn.EfficientNetB0(include_top=False, weights='imagenet', input_shape=input_shape) File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/efficientnet/init.py", line 44, in wrapper return func(*args, **kwargs) File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/efficientnet/model.py", line 477, in EfficientNetB0 **kwargs File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/efficientnet/model.py", line 326, in EfficientNet weights=weights) File "/home/mbarakat/anaconda3/envs/tf1/lib/python3.6/site-packages/keras_applications/imagenet_utils.py", line 316, in _obtain_input_shape 'input_shape=' + str(input_shape) + '') ValueError: The input must have 3 channels; got input_shape=(256, 256, 1)

[innat]

You can add a convolutional layer with feature maps 3 with pad size 3 (same) to match with the imagenet weights.

[atamazian]

This issue is not related to the package itself, so I suggest to close it.