• Stars
    star
    188
  • Rank 205,563 (Top 5 %)
  • Language
    Python
  • License
    Apache License 2.0
  • Created almost 2 years ago
  • Updated over 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

aim-mlflow integration

aimlflow

Aim-powered supercharged UI for MLFlow logs

Run beautiful UI on top of your MLflow logs and get powerful run comparison features.

Platform Support PyPI - Python Version PyPI Package License


About โ€ข Getting Started โ€ข Why use aimlflow? โ€ข Use Cases โ€ข Blog

โ„น๏ธ About

aimlflow helps to explore various types of metadata tracked during the training with MLFLow, including:

  • hyper-parameters
  • metrics
  • images
  • audio
  • text

More about Aim: https://github.com/aimhubio/aim

More about MLFLow: https://github.com/mlflow/mlflow

๐Ÿ Getting Started

Follow the steps below to set up aimlflow.

  1. Install aimlflow on your training environment:
pip install aim-mlflow
  1. Run live time convertor to sync MLFlow logs with Aim:
aimlflow sync --mlflow-tracking-uri={mlflow_uri} --aim-repo={aim_repo_path}
  1. Run the Aim UI:
aim up --repo={aim_repo_path}

๐Ÿ”ฆ Why use aimlflow?

  1. Powerful pythonic search to select the runs you want to analyze.

image

  1. Group metrics by hyperparameters to analyze hyperparametersโ€™ influence on run performance.

image

  1. Select multiple metrics and analyze them side by side.

image

  1. Aggregate metrics by std.dev, std.err, conf.interval.

image

  1. Align x axis by any other metric.

image

  1. Scatter plots to learn correlations and trends.

image

  1. High dimensional data visualization via parallel coordinate plot.

image

๐ŸŽฌ Use Cases

๐ŸŽ‡ Read the article: Exploring MLflow experiments with a powerful UI

image

๐Ÿ” Read the article: How to integrate aimlflow with your remote MLflow

image

๐Ÿ“Š Read the article: Aim and MLflow โ€” Choosing Experiment Tracker for Zero-Shot Cross-Lingual Transfer

image

More questions?

  1. Read the docs
  2. Open a feature request or report a bug
  3. Join Discord community server