Skip to content

output of step3 (Run data pipeline) #4

@ljjcoder

Description

@ljjcoder

Hello, when I run step3 (sh run_data_pipeline.sh), the output json file content is as follows:

Image The process supervision tags are not extracted at all.

After inspecting the code, I found that the output of model_outputs = self.model.generate(inputs, sampling_params=sampling_params) did not follow the self.pattern2 pattern, resulting in matches = re.findall(self.pattern2, response, re.DOTALL) being empty.

However, I checked the code and did not find any instructions indicating that the output of the model needs to follow self.pattern2. I'm not sure if I've missed a crucial step that's preventing me from generating the correct data. I'd appreciate your help.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions