Skip to main content
Ctrl+K
AMD Logo
Quark Version List
  • GitHub
  • Support

Quark 0.6.0 documentation

Release Notes

  • Release Information

Getting Started with Quark

  • Introduction to Quantization
  • Installation
  • Basic Usage
    • Quark for PyTorch
    • Quark for ONNX
  • Accessing PyTorch Examples
  • Accessing ONNX Examples

Advanced Quark Features for PyTorch

  • Configuring PyTorch Quantization
  • Calibration Methods
  • Calibration Datasets
  • Quantization Strategies
  • Quantization Schemes
  • Quantization Symmetry
  • Language Model Optimization
    • Pruning
    • Language Model Post Training Quantization (PTQ) Using Quark
    • Language Model QAT Using Quark
    • Language Model Evaluation in Quark
  • Exporting Quantized Models
    • Bridge from Quark to llama.cpp
  • Exporting Using ONNX Runtime Gen AI Model Builder
  • Extensions
    • Integration with AMD Pytorch-light (APL)
    • Brevitas Integration
  • Using MX (Microscaling)
  • Two Level Quantization Formats

Advanced Quark Features for ONNX

  • Configuring ONNX Quantization
  • Calibration methods
  • Calibration datasets
  • Quantization Strategies
  • Quantization Schemes
  • Quantization Symmetry
  • Mixed Precision
  • Block Floating Point 16 (BFP16)
  • Microscaling (MX)
  • Accuracy Improvement Algorithms
    • Quantizing Using CrossLayerEqualization (CLE)
    • Quantization Using AdaQuant and AdaRound
    • SmoothQuant (SQ)
    • Quantizating a model with GPTQ
  • Optional Utilities
  • Tools

APIs

  • PyTorch APIs
    • Pruning
    • Quantization
    • Export
    • Pruner Configuration
    • Quantizer Configuration
    • Exporter Configuration
  • ONNX APIs
    • Quantization
    • Optimization
    • Calibration
    • ONNX Quantizer
    • QDQ Quantizer
    • Configuration
    • Quantization Utilities

Troubleshooting and Support

  • PyTorch FAQ
  • ONNX FAQ
  • quark.onnx.tools

quark.onnx.tools

Contents

  • Submodules

quark.onnx.tools#

Submodules#

  • quark.onnx.tools.convert_a8w8_npu_to_a8w8_cpu
  • quark.onnx.tools.convert_customqdq_to_qdq
  • quark.onnx.tools.convert_dynamic_to_fixed
  • quark.onnx.tools.convert_fp16_to_fp32
  • quark.onnx.tools.convert_fp32_to_fp16
  • quark.onnx.tools.convert_lstm_to_customlstm
  • quark.onnx.tools.convert_nchw_to_nhwc
  • quark.onnx.tools.convert_onnx_to_onnxtxt
  • quark.onnx.tools.convert_onnxtxt_to_onnx
  • quark.onnx.tools.convert_qdq_to_qop
  • quark.onnx.tools.convert_quant_to_float
  • quark.onnx.tools.convert_resize_fs_to_pof2s
  • quark.onnx.tools.convert_s8s8_to_u8s8
  • quark.onnx.tools.convert_u16s8_to_s16s8
  • quark.onnx.tools.convert_u16u8_to_u8u8
  • quark.onnx.tools.print_a16w8_a8w8_nodes
  • quark.onnx.tools.remove_initializer_from_input
  • quark.onnx.tools.remove_qdq
  • quark.onnx.tools.remove_qdq_between_ops
  • quark.onnx.tools.remove_qdq_mul_add
  • quark.onnx.tools.save_tensor_hist
  • quark.onnx.tools.save_weights_hist
Contents
  • Submodules

Last updated on Dec 10, 2024.

  • Terms and Conditions
  • Quark Licenses and Disclaimers
  • Privacy
  • Trademarks
  • Statement on Forced Labor
  • Fair and Open Competition
  • UK Tax Strategy
  • Cookie Policy
  • Cookie Settings
© 2024 Advanced Micro Devices, Inc