Possibility to skip optional compression ratio

[hugoabonizio]

Since the compression ratio is optional information, can we have an option to skip its calculation in ‎QuantizationConfig.from_pretrained? It takes a very long time to run for large models.

compressed-tensors/src/compressed_tensors/quantization/quant_config.py

Line 235 in b8cf630

compression_ratio = calculate_compression_ratio(model)