Awesome
<div align="center"> <table> <tbody> <tr> Powered by Aim <td> <a href="https://github.com/aimhubio/aim">Drop a star to support Aim ⭐</td> <td> <a href="https://community.aimstack.io/">Join Aim discord community</a> <img width="20px" src="https://user-images.githubusercontent.com/13848158/226759622-063b725d-8b3e-4c75-80c7-11fb04b7adf5.png" </tr> </tbody> </table> </div> <div align="center"> <h1>aimlflow</h1> <h3>Aim-powered supercharged UI for MLFlow logs</h3> Run beautiful UI on top of your MLflow logs and get powerful run comparison features. </div> <br/> <div align="center">
ℹ️ 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
🏁 Getting Started
Follow the steps below to set up aimlflow.
- Install aimlflow on your training environment:
pip install aim-mlflow
- Run live time convertor to sync MLFlow logs with Aim:
aimlflow sync --mlflow-tracking-uri={mlflow_uri} --aim-repo={aim_repo_path}
- Run the Aim UI:
aim up --repo={aim_repo_path}
🔦 Why use aimlflow?
- Powerful pythonic search to select the runs you want to analyze.
- Group metrics by hyperparameters to analyze hyperparameters’ influence on run performance.
- Select multiple metrics and analyze them side by side.
- Aggregate metrics by std.dev, std.err, conf.interval.
- Align x axis by any other metric.
- Scatter plots to learn correlations and trends.
- High dimensional data visualization via parallel coordinate plot.
🎬 Use Cases
🎇 Read the article: Exploring MLflow experiments with a powerful UI
🔍 Read the article: How to integrate aimlflow with your remote MLflow
📊 Read the article: Aim and MLflow — Choosing Experiment Tracker for Zero-Shot Cross-Lingual Transfer
