Register
Login
Resources
Docs Blog Datasets Glossary Case Studies Tutorials & Webinars
Product
Data Engine LLMs Platform Enterprise
Pricing Explore
Connect to our Discord channel
..
fe46adaa97
Initial commit
4 years ago
9e74ba23d5
Higher LR and batch size
4 years ago
fe46adaa97
Initial commit
4 years ago
9e74ba23d5
Higher LR and batch size
4 years ago
82b24d5f99
Unmark node as metric
4 years ago

README.md

You have to be logged in to leave a comment. Sign In

Hyperparameters as DVC dependency

This example how to set up your hyperparameter file as a DVC dependency of the training stage.
This means that you manually edit your params.yml file before training, then use dvc repro to run the training stage. In theory, this is the correct workflow with DVC.

The relevant files are:

  • mnist_trainer.py - the main script, defines the pytorch-lightning Trainer and connects it to the DAGsHubLogger.
  • params.yml - Hyperparameters to be used for the training run. To try an experiment with a different hyperparameter configurations, you should edit this file before running dvc repro.
    This YAML format is supported by DAGsHub, for smart experiment comparison and display.
  • metrics.csv - Output metrics from the training stage.
    This CSV format is supported by DAGsHub, for smart experiment comparison and display.
  • train.dvc - The DVC stage file.

Made with 🐶 by DAGsHub.

Tip!

Press p or to see the previous file or, n or to see the next file

Comments

Loading...