Skip to content

Questions Regarding Inconsistent Evaluation Metrics and Cross-Validation Implementation #55

Description

@xiomg666

Hello, I hope this message finds you well. I am currently working with your CLEAN model for protein sequence inference and have a couple of questions I would greatly appreciate your insights on.

  1. Inconsistent Evaluation Metrics
    I have noticed that when I run the inference with the same test dataset and the same model file, the evaluation metrics (AUC, Recall, Precision, F1) vary with each run. This occurs even though the input data and model weights remain the same. Could you help me understand what might be causing this inconsistency? Are there any sources of randomness in the inference pipeline that could be affecting the results?

  2. 5-Fold Cross-Validation Implementation
    I am also interested in understanding how the 5-fold cross-validation is implemented in the original code. Could you please provide more details or point me to the part of the code where this implementation occurs? I would appreciate any clarification on how the data is split and how the cross-validation process is integrated with the inference and model evaluation.

Thank you for your time and for creating such a useful tool. I look forward to your response!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions