PEFT documentation
Errors
Get started
Guides
Configurations and modelsIntegrations Memory Efficient TrainingModel mergingQuantizationCustom modelsAdapter injectionMixing PEFT methodstorch.compileContribute to PEFTTroubleshootingPEFT checkpoint format
Distributed Training
Methods
API reference
You are viewing main version, which requires installation from source. If you'd like
regular pip install, checkout the latest stable version (v0.20.0).
Errors
Base PEFT error
The adapter being injected matched no module or parameter of the base model.
Raised e.g. when the target_modules of the PEFT config matched nothing, which most often points at a
misconfiguration. It subclasses ValueError for backwards compatibility with code that intercepted the generic
error raised previously. Code that adds such an adapter intentionally can catch this error (or PeftError) and
proceed.