A collection of INC3-related PRs

[yiliu30]

• INC 3.X

  • https://github.com/intel/neural-compressor/pulls?q=is%3Apr+label%3AINC3.X+

• Auto-Tune

  • https://github.com/intel/neural-compressor/pulls?q=is%3Apr+label%3AAuto-Tune+label%3AINC3.X+

• Torch

  • https://github.com/intel/neural-compressor/pulls?q=is%3Apr+label%3AINC3.X+label%3APyTorch+

• Onnx Runetime

  • https://github.com/intel/neural-compressor/pulls?q=is%3Apr+label%3AINC3.X+label%3A%22Onnx+Runtime%22

• TF

  • https://github.com/intel/neural-compressor/pulls?q=is%3Apr+label%3AINC3.X+label%3ATensorflow+

[yiliu30]

Until January 2024:

• Common

  • refine release binary build for 3.0 API #1336
  • Introduce INC 3.0 quantization API and port torch RTN into 3.0 #1380
  • Enhance 3.x API #1397
  • Add regular expression matching for base config #1510
  • Add Config Registry for autotune #1543
  • Refactor the tuning pipeline #1528
  • Rename RTNWeightOnlyConfig to RTNConfig #1551
  • Add default workspace for 3.x #1553
  • Add default config set for tuning #1562

• Torch

  • Port torch GPTQ to 3.x #1408
  • PyTorch AWQ Weight-only 3x API Implementation  #1561
  • Update publication_list.md #1567

• ORT

  • Migrate onnxrt RTN WOQ to 3.x API #1544
  • Fix common import for onnxrt #1559
  • Migrate ONNXRT SmoothQuant to 3.x  #1565
  • Migrate onnxrt GPTQ and AWQ WOQ to 3.x API #1570
  • Enable LWQ for onnxrt WOQ in 3.x API #1625

• TF

  • Update INC 3.0 New API for Keras Static Quant #1400