LightningToolbox: A PyTorch Lightning Facilitator
Installation • Docs • License
Welcome to lightning-toolbox, a python package that offers a set of automation tools built on top of PyTorch Lightning. As a deep learning researcher, I found that PyTorch Lightning offloads a significant portion of redundant work. However, I still found myself spending a considerable amount of time on writing boilerplate code for datamodules, training/validation steps, logging, and specially for not so complicated costum training loops. This is why I created lightning-toolbox - to make it easier to focus on writing experiment-specific code rather than dealing with tedious setup tasks.
By passing your PyTorch model onto a generic lightning.LightningModule (lightning_toolbox.TrainingModule), lightning-toolbox automatically populates the objective function, optimizer step, and more. In addition, lightning-toolbox's generic lightning.LightningDataModule (lightning_toolbox.DataModule) can turn any PyTorch dataset, into a experiment-ready lightning data module, completing the cycle for writing lightning deep learning experiments.
Most of the functionality provided in this package is based on dypy, which enables lazy evaluation of variables and runtime code injections. Although lightning-toolbox is currently in its early stages and mainly serves as a facilitator for my personal research projects, I believe it can be helpful for many others who deal with similar deep learning experiments. Therefore, I decided to open-source this project and continue to add on to it as I move further in my research.
As a disclaimer this package does not intend to solve field-specific problems and provides more generic facilitator codes that the official lightning-bolts, that you can easily mold into your desired deep learning experiment.
Installation
pip install lightning-toolbox
Lightning toolbox is tested on lightning==1.9.0, although there's no version restriction setup for this package, things might break down if the community decides to roll backward incompatible changes to the core Pytorch Lightning API (as they usually do).
License
This project is licensed under the terms of the Apache 2.0 license. See LICENSE for more details.
1 maintainer
Metadata
1 maintainer
Release files for lightning-toolbox 0.0.35
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| lightning_toolbox-0.0.35.tar.gz | 32.8 kB | Without provenance | Details for lightning_toolbox-0.0.35.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| lightning_toolbox-0.0.35-py3-none-any.whl | Python 3 | none | any | Details for lightning_toolbox-0.0.35-py3-none-any.whl |
Total release size: 79.4 kB
Release files / lightning_toolbox-0.0.35.tar.gz
| Download URL | lightning_toolbox-0.0.35.tar.gz |
|---|---|
| Size | 32.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f698ffeb9685a89cdfbbef78d4bbec7b4cb13dacaea78ba66ec0e11278a0f140
|
|
BLAKE2b-256 checksum How to use checksums |
9f783767ade9a3a3c858d38024e6f86410937a5f6261d08544fd2fbf46496d96
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.8.18
|
Release files / lightning_toolbox-0.0.35-py3-none-any.whl
| Download URL | lightning_toolbox-0.0.35-py3-none-any.whl |
|---|---|
| Size | 46.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
90c19173dfe5572f46dc5b666678b2e93a9e3c16af0b8f29f83a726d540ac2d9
|
|
BLAKE2b-256 checksum How to use checksums |
7d7cc36af7e2778dc27a5c1076d6c1a625e5924890005ef02b1d41346579cef2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.8.18
|
Security for lightning-toolbox 0.0.35
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
