June 14, 2023

Effortlessly analyze the performance, accuracy, and cost of training your machine learning model by comparing different experiments.

Gain insights into the package requirements necessary to run an experiment.

Additionally, you now have the ability to create a virtualenv or a conda-env using these requirements.

virtualenv

$ markov experiment virtualenv -e <experiment_id>

conda

$ markov experiment conda-env -e <experiment_id>

SDK has been updated to version 1.3.3

Major updates in the SDK include:

Summary feature: Add a summary to your running or completed experiment, allowing for a quick overview and easier comparison of experiments. Find the instructions here.
LightGBM integration: Track experiments created with lightgbm using markov.lightgbm.auto_record() .

Please note that you will need to update your SDK. Please follow the instructions to update your SDK installation here.

Last updated 2 years ago