Making the Beby-Gan model quite a bit smaller.

quant.py

# Before running this, setup a python venv for everything:
# python3 -m venv coremltools-venv
# source coremltools-venv/bin/activate
# pip install -U coremltools

import coremltools as ct
from coremltools.models.neural_network import quantization_utils

# load full precision model
model_fp32 = ct.models.MLModel('Beby-GANx4.mlmodel')

model_fp16 = quantization_utils.quantize_weights(model_fp32, nbits=16)
model_fp16.save("Beby-GANx4fp16.mlmodel")

model_fp8 = quantization_utils.quantize_weights(model_fp32, nbits=8)
model_fp8.save("Beby-GANx4fp8.mlmodel")