mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 00:21:07 +01:00
9944cd0949
Related to #155032 - ✅ quantization-accuracy-debugging.rst: [Preview](https://docs-preview.pytorch.org/pytorch/pytorch/155520/quantization-accuracy-debugging.html) vs [main](https://docs.pytorch.org/docs/main/quantization-accuracy-debugging.html) - ✅ quantization-backend-configuration.rst: [Preview](https://docs-preview.pytorch.org/pytorch/pytorch/155520/quantization-backend-configuration.html) vs [main](https://docs.pytorch.org/docs/main/quantization-backend-configuration.html) - ✅ quantization-support.rst: [Preview](https://docs-preview.pytorch.org/pytorch/pytorch/155520/quantization-support.html) vs [main](https://docs.pytorch.org/docs/main/quantization-support.html) - ✅ random.rst: [Preview](https://docs-preview.pytorch.org/pytorch/pytorch/155520/random.html) vs [main](https://docs.pytorch.org/docs/main/random.html) Pull Request resolved: https://github.com/pytorch/pytorch/pull/155520 Approved by: https://github.com/svekars Co-authored-by: Svetlana Karslioglu <svekars@meta.com>
20 lines
553 B
Markdown
20 lines
553 B
Markdown
# Quantization Backend Configuration
|
|
|
|
FX Graph Mode Quantization allows the user to configure various
|
|
quantization behaviors of an op in order to match the expectation
|
|
of their backend.
|
|
|
|
In the future, this document will contain a detailed spec of
|
|
these configurations.
|
|
|
|
## Default values for native configurations
|
|
|
|
Below is the output of the configuration for quantization of ops
|
|
in x86 and qnnpack (PyTorch's default quantized backends).
|
|
|
|
Results:
|
|
|
|
```{eval-rst}
|
|
.. literalinclude:: scripts/quantization_backend_configs/default_backend_config.txt
|
|
```
|