ValueError: Shapes are incompatible when saving and loading model using Subpixel

[adrigrillo]

Generating the model that uses the Subpixel layer produces an error when the loading from a saved checkpoint. The problem is located in this line:

Keras-segmentation-deeplab-v3.1/subpixel.py

Line 101 in 2138ad3

config['filters']= int(config['filters'] / self.r*self.r)

That can be solved just putting some parenthesis wrapping the r product like this:

config['filters']= int(config['filters'] / (self.r*self.r))

Additionally, using tensorflow 1.14 and tf.keras included in mentioned version the same method raises and error as the key rank does not exist in the Conv2D layer. The problematic line is:

Keras-segmentation-deeplab-v3.1/subpixel.py

Line 99 in 2138ad3

config.pop('rank')