**Feat: Integrate Loss Graphs into Gradio Interface** #2241
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
**Body:** * Implements helper functions to retrieve and display loss graphs for different projects. * Integrates with Gradio to provide interactive tabs with loss graph images. * Handles project selection and updates loss graphs accordingly. * Includes helper functions for generating and saving loss graphs from TensorBoard logs
- Updated `extract_scalar_data` to handle NaN values in TensorBoard logs. - If a scalar value is NaN, the method now falls back to the previous valid value. - If no previous value is available, a default of 0.0 is used. - This ensures continuity and robustness in the extracted scalar data.
…ault project name, also moved default_project_name logic into the get_projects function.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Pull Request Checklist
PR Type
Description
This pull request introduces a new feature to enhance the Gradio interface by adding the ability to visualize loss graphs:
Screenshot