trail demo library
Project description
Trail
Trail brings more transparency in your ml experimentation. Start by using mlflow to track experiments and follow the steps below.
Installation
Install Trail from Pypi via '''pip install traildb'''
Get started
''' from traildb import trail_init '''
Initialize a trail object
add this line of code in the beginning of the trainingscript.
''' trail = trail_init(username, password) '''
The input paramter "username" and "password" will be provided by the trail-team.
log experiment
Call the log_experiment() method after the mlflow run (not within the run)
''' with mlflow.start_run() as run:
...your training code...
trail.log_experiment(mlflow.get_run(run_id=run.info.run_id), parent_id:"String", data_meta:{dict}) '''
The input paramters parent_id [String] and data_meta [dict] must be provided in the according type. If they are non existing please provide empty String ("") or dict ({})
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
