pytorch / pytorch/vision

Add argument to checkpoint every *n* epochs in references/classification/train.py

Open
#7,481 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
17.9k
Forks
7.3k
Avg merge
1d 15h
Merged PRs (30d)
13

Description

🚀 The feature

In the training script mentioned in the title, the model + training is checkpointed after each epoch if --output-dir is provided.
I would like to add an argument which gives the option of checkpointing every n epochs.,

Motivation, pitch

As a DL researcher, I am using this script for training several models on ImageNet. As I don't want to checkpoint after each epoch I would like more flexibility.

Alternatives

No response

Additional context

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in references/classification/train.py and inspect how --output-dir currently controls per-epoch checkpointing. Add a command-line option for the checkpoint interval, then verify that checkpoints occur at the requested epoch cadence while preserving the existing behavior when the option is not used.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.