notes

Contents:

  • C++
  • Sphinx
  • git
  • docker
  • LaTeX
  • Kaldi
  • bash
  • CUDA
  • torch
    • torch.load and torch.save
    • torch.gather
    • DDP
    • TorchScript
    • Logical operations
    • Note
    • Quantization
      • Internals
      • torch.quantize_per_tensor
      • quantize_per_tensor_dynamic
      • torch.quantize_per_channel
      • Observer
      • Hello
      • References
    • android
    • onnx
    • nn.LSTM
    • torchvision
    • audograd
    • seed
    • float16
    • Dynamo
    • random
    • compile
    • benchmark
    • cv2
    • einsum
  • Python
  • java
  • javascript
  • HTML
  • css
  • pybind11
  • Protocol Buffers
  • gRPC
  • lwn.net
  • Linker and Loader
  • espnet
  • cmake
  • huggingface
  • EECS E6870 Speech Recognition
  • ncnn
  • LLVM
  • Android
  • qemu
  • sox
  • MNN
  • SIMD
  • asio
  • websocketpp
  • Operating systems
  • encoding
  • ios
  • Embedded systems
  • ssh
  • onnx
  • csharp
  • Flask
  • ARM
  • VirtualBox
  • Go
  • Whisper
  • Windows
  • qt
  • webassembly
  • spleeter
  • django
  • React
  • tts
  • rust
  • ELF
  • ROS2
  • OpenFst
  • Colab
  • Dart
  • Flutter
  • Keyword spotting (KWS)
  • Papers
  • Pascal
  • ggml
  • Amphion
  • HarmonyOS
  • icefall
  • RKNN
  • lhotse
  • ffmpeg
  • vlc
notes
  • torch
  • Quantization
  • Edit on GitHub

Quantization

  • Internals
    • QScheme
    • PerTensorAffineQuantizer
  • torch.quantize_per_tensor
    • Compress ration
  • quantize_per_tensor_dynamic
  • torch.quantize_per_channel
  • Observer
  • Hello
  • References
    • Links
Previous Next

© Copyright 2022, fangjun.

Built with Sphinx using a theme provided by Read the Docs.