fastvideo.v1.layers.quantization

`fastvideo.v1.layers.quantization`#

Submodules#

fastvideo.v1.layers.quantization.base_config

Package Contents#

Functions#

`get_quantization_config`
`register_quantization_config`	Register a customized vllm quantization config.

Data#

`QUANTIZATION_METHODS`
`QuantizationMethods`
`all`

API#

fastvideo.v1.layers.quantization.QUANTIZATION_METHODS: list[str][source]#: ‘list(…)’

fastvideo.v1.layers.quantization.QuantizationMethods[source]#: None

fastvideo.v1.layers.quantization.all[source]#: [‘QuantizationMethods’, ‘QuantizationConfig’, ‘get_quantization_config’, ‘QUANTIZATION_METHODS’]

fastvideo.v1.layers.quantization.get_quantization_config(quantization: str) → type[fastvideo.v1.layers.quantization.base_config.QuantizationConfig][source]#

fastvideo.v1.layers.quantization.register_quantization_config(quantization: str)[source]#

Register a customized vllm quantization config.

When a quantization method is not supported by vllm, you can register a customized quantization config to support it.

Parameters:: quantization (str) – The quantization method name.

.. rubric:: Examples

from fastvideo.v1.layers.quantization import register_quantization_config from fastvideo.v1.layers.quantization import get_quantization_config from fastvideo.v1.layers.quantization.base_config import QuantizationConfig

@register_quantization_config(“my_quant”) … class MyQuantConfig(QuantizationConfig): … pass

get_quantization_config(“my_quant”) <class ‘MyQuantConfig’>