The TensorFlow Cloud repository provides APIs that will allow to easily go from debugging and training your Keras and TensorFlow code in a local environment to distributed training in the cloud.
up before the "pip install" block.
This way pip will see a custom wheel file that is lies next to the entry point and can be installed using a line in the requirements.txt file.
Suggestion for easy patch to support custom wheel files. Please move in containerize.py the line
lines.append("COPY {} {}".format(self.destination_dir, self.destination_dir))up before the "pip install" block. This way pip will see a custom wheel file that is lies next to the entry point and can be installed using a line in the requirements.txt file.
See stackoverflow question: https://stackoverflow.com/questions/70249688/how-to-use-tensorflow-cloud-with-my-own-python-wheel