Extracting features from ResNet101

[fmaglia]

Hi,

I'm trying to extract features from the last convolutional layer of ResNet101, from the image at its original dimension.
I obtain this error:
ValueError: The shape of the input to "Flatten" is not fully defined (got (None, None, 2048). Make sure to pass a complete "input_shape" or "batch_input_shape" argument to the first layer in your model.
Here is my call:
base_model = resnet101_model(None, None, 3, 0)
Can you help me?