Cadene / Bootstrap.pytorch

bootstrap.pytorch is a high-level extension for deep learning projects with PyTorch. It aims at accelerating research projects and prototyping by providing a powerful workflow focused on your dataset and model.

And it is:

• Scalable
• Modular
• Shareable
• Extendable
• Uncomplicated
• Built for reproducibility
• Easy to log and plot anything

Unlike many others, bootstrap.pytorch is not a wrapper over pytorch, but a powerful extension.

Quick tour

To run an experiment (training + evaluation):

python -m bootstrap.run
       -o myproject/options/sgd.yaml

To display parsed options from the yaml file:

python -m bootstrap.run
       -o myproject/options/sgd.yaml
       -h

Running an experiment will create 4 files, here is an example with mnist:

• options.yaml contains the options used for the experiment
• logs.txt contains all the information given to the logger
• logs.json contains the following data: train_epoch.loss, train_batch.loss, eval_epoch.accuracy_top1, etc
• view.html contains training and evaluation curves with javascript utilities (plotly)

To save the next experiment in a specific directory:

python -m bootstrap.run
       -o myproject/options/sgd.yaml
       --exp.dir logs/custom

To reload an experiment:

python -m bootstrap.run
       -o logs/custom/options.yaml
       --exp.resume last

Documentation

The package reference is available on the documentation website.

It also contains some notes:

• Installation
• Concepts
• Quickstart
• Directories
• Examples

Official project modules

• mnist.bootstrap.pytorch is a useful example for starting a quick project with bootstrap
• vision.bootstrap.pytorch contains utilities to train image classifier, object detector, etc. on usual datasets like imagenet, cifar10, cifar100, coco, visual genome, etc
• recipe1m.bootstrap.pytorch is a project for image-text retrieval related to the Recip1M dataset developped in the context of a SIGIR18 paper
• block.bootstrap.pytorch is a project focused on fusion modules related to the VQA 2.0, TDIUC and VRD datasets developped in the context of a AAAI19 paper

Poster

Contribute

Contributions to this repository are welcome and encouraged. We also have a public trello board with prospect features, as well as an indication of those currently being developed. Feel free to contact us with suggestions, or send a pull request.

We use flake8 to perform early semantic checking of submitted code. After installing all the requirements in requirements.txt, please run the following to activate the pre-commit hooks for flake8: pre-commit install

To manually trigger the pre-commit checks for a file without creating a commit, you can run the following command: pre-commit run --files <your_files.py>

bootstrap.pytorch was conceived and is maintained by Rémi Cadène and Micael Carvalho, with helpful discussions and insights from Thomas Robert and Hedi Ben-Younes. We chose to adopt the [very permissive] BSD-3 license, which allows for commercial and private use, making it compatible with both academy and industry standards.