Skip to content

Conversation

@Gautam-Galada
Copy link
Contributor

Description

This pull request adds visualization feature with 6 plots currently , keeping the current MLX training setup in mind. The image contents are as follows :

models/visualization/{timestamp}/plots/
├── 1_loss_curves.png
├── 2_learning_rate.png
├── 3_training_speed.png
├── 4_throughput.png
├── 5_memory_usage.png
└── 6_token_progress.png

How to Test

# Default: Creates 6 separate files
python scripts/visualize_mlx_training.py

# Legacy: Single combined plot
python scripts/visualize_mlx_training.py --combined

# High resolution
python scripts/visualize_mlx_training.py --dpi 300

# Custom directory
python scripts/visualize_mlx_training.py --output my_analysis

# Custom timestamp
python scripts/visualize_mlx_training.py --timestamp 2026-02-01_01-28-37

Checklist

Added matplotlib to pyproject.toml
Unit-testing is included by pytest tests/test_visualize_mlx_training.py -v

Additional Information

The visualization code is folder friendly, it creates the plots for the latest timestamp, or specific if provided as an argument.

Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

This pull request adds a comprehensive visualization feature for MLX training logs. The implementation includes a script that parses training logs, extracts various metrics (loss, learning rate, throughput, memory usage, etc.), and generates informative plots to help analyze training runs.

Changes:

  • Added scripts/visualize_mlx_training.py with log parsing, run management, and visualization capabilities
  • Added unit tests for log parsing and run management functionality
  • Added matplotlib as a dependency in pyproject.toml

Reviewed changes

Copilot reviewed 3 out of 4 changed files in this pull request and generated 4 comments.

File Description
scripts/visualize_mlx_training.py New script implementing MLX training log parser, run manager, and visualizer with support for both individual and combined plot generation
tests/test_visualize_mlx_training.py Unit tests for log parsing and run management functionality
pyproject.toml Added matplotlib dependency for visualization support

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Gautam-Galada and others added 4 commits February 2, 2026 14:59
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
@Sudhendra Sudhendra merged commit 6bf3b73 into Sudhendra:main Feb 2, 2026
0 of 6 checks passed
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.

2 participants