auto-round
refactor eval and add UT
#1324
Merged

Commits
  • refactor eval and add UT
    root committed 70 days ago
  • allow_deprecated_quantization and simplify UT to reduce time
    xin3he committed 69 days ago
  • fix CI and update readme
    xin3he committed 69 days ago
  • remove 'from auto_round import AutoRoundConfig'
    xin3he committed 69 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 69 days ago
  • Update test/test_cuda/advanced/test_evaluation.py
    xin3he committed 69 days ago
  • remove AutoHfQuantizer
    xin3he committed 69 days ago
  • Refactor imports in evaluation.py and update logging in eval_cli.py
    xin3he committed 69 days ago
  • Update evaluation commands to disable optimization return and increase limit
    xin3he committed 69 days ago
  • Add tensor_parallel_size handling and CUDA_VISIBLE_DEVICES setup in eval_with_vllm
    xin3he committed 69 days ago
  • Update step-by-step documentation for quantization and evaluation processes
    xin3he committed 69 days ago
  • Fix task handling in eval_with_vllm to support comma-separated task strings
    xin3he committed 69 days ago
  • Update vllm_args help text and normalize argument format in parse_vllm_args
    xin3he committed 69 days ago
  • Refactor imports to include AutoRoundConfig in test files
    xin3he committed 68 days ago
  • support general devices
    xin3he committed 66 days ago
  • Merge branch 'main' into xinhe/eval
    xin3he committed 66 days ago
  • Update environment variable for XPU in eval_with_vllm function
    xin3he committed 66 days ago
  • Merge branch 'main' into xinhe/eval
    xin3he committed 66 days ago
  • fix CI
    xin3he committed 66 days ago
  • fix test_cuda failure
    xin3he committed 66 days ago
Loading