Skip to content

Modify pipeline and automap#16907

Closed
RkGrit wants to merge 2 commits intoapache:masterfrom
RkGrit:modify_pipeline_and_automap
Closed

Modify pipeline and automap#16907
RkGrit wants to merge 2 commits intoapache:masterfrom
RkGrit:modify_pipeline_and_automap

Conversation

@RkGrit
Copy link
Contributor

@RkGrit RkGrit commented Dec 15, 2025

Two modifications regarding model management, including:

  • Expose the preprocess and postprocess interfaces of the Pipeline class.
  • Delete config_cls and model_cls in ModelInfo, only maintain auto_map for hf transformers model.

@RkGrit RkGrit force-pushed the modify_pipeline_and_automap branch from ea7e672 to c4a9866 Compare December 15, 2025 15:01
@yunbow30944
Copy link
Contributor

LGTM!

Copy link
Contributor

@CRZbulabula CRZbulabula left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

PTAL.

Comment on lines +51 to +58
def preprocess(self, inputs):
return inputs

@abstractmethod
def forecast(self, inputs, **infer_kwargs):
pass

def _postprocess(self, output: torch.Tensor):
def postprocess(self, output: torch.Tensor):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

U should add some annotations to notify the shape of inputs and output for forecasting task.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And the shape of inputs should be 3-d.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants