fix: Always return a list of quantization bits values from `get_quantization`

ludwig-ai / ludwig

Low-code framework for building custom LLMs, neural networks, and other AI models

http://ludwig.ai

Apache License 2.0

11.21k stars 1.19k forks source link

fix: Always return a list of quantization bits values from `get_quantization` #3926

Closed jeffkinnison closed 9 months ago

jeffkinnison commented 9 months ago

ludwig.utils.config_utils.get_quantization returns the value of quantization.bits for LLM model types (either int or None type), but for ECD it returns a list of values to account for multiple LLM encoders. To simplify checking the quantization, get_quantization now returns a list in every case.

github-actions[bot] commented 9 months ago

Unit Test Results

  4 files ±0   4 suites ±0 9m 25s :stopwatch: -17s 12 tests ±0   9 :heavy_check_mark: ±0   3 :zzz: ±0 0 :x: ±0 40 runs ±0 28 :heavy_check_mark: ±0 12 :zzz: ±0 0 :x: ±0

Results for commit b958a34c. ± Comparison against base commit 51f38c5c.

:recycle: This comment has been updated with latest results.