Skip to content

Latest commit

 

History

History
24 lines (17 loc) · 786 Bytes

training_notebook_runbook.md

File metadata and controls

24 lines (17 loc) · 786 Bytes

Training notebook runbook

Usage

Here are instructions for training a model using the Colab notebook script.

Repo clone config

At the Clone repo section, put in GitHub user and access token in indicated locations.

Modify hyperparameters for training

In the first few lines after Train transformer, there are hyperparameters that can be changed for training, including epochs and batch size.

Authorize google account

When the first cell is run, follow the prompt for authorization.

Features

The Colab script does the following:

  • Copy dataset from gcp bucket to Colab environment
  • Clone repo
  • Train model
  • Save checkpoints on gcp bucket while training
  • Evaluate every checkpoint after training
  • Update desired google sheet with evaluation results