Metadata-Version: 2.4
Name: torchmil
Version: 0.1.0
Summary: Deep Multiple Instance Learning library for Pytorch
Author-email: Francisco Miguel Castro-Macías <francastro8b@gmail.com>
License-Expression: Apache-2.0
Project-URL: homepage, https://franblueee.github.io/torchmil/
Project-URL: documentation, https://franblueee.github.io/torchmil/api/
Project-URL: repository, https://github.com/Franblueee/torchmil
Project-URL: issues, https://github.com/Franblueee/torchmil/issues
Keywords: deep-learning,pytorch,multiple-instance-learning,multi-instance-learning,mil
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Information Technology
Classifier: Intended Audience :: Science/Research
Classifier: Natural Language :: English
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Classifier: Topic :: Scientific/Engineering :: Mathematics
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch
Requires-Dist: tensordict
Requires-Dist: torchmetrics
Requires-Dist: einops
Requires-Dist: scipy
Requires-Dist: tqdm
Requires-Dist: numpy
Provides-Extra: docs
Requires-Dist: mkdocs; extra == "docs"
Requires-Dist: mkdocs-material; extra == "docs"
Requires-Dist: mknotebooks; extra == "docs"
Requires-Dist: mkdocstrings; extra == "docs"
Requires-Dist: mkdocstrings-python; extra == "docs"
Requires-Dist: mkdocs-exclude; extra == "docs"
Provides-Extra: tests
Requires-Dist: torchmil; extra == "tests"
Requires-Dist: pytest; extra == "tests"
Requires-Dist: pytest-cov; extra == "tests"
Requires-Dist: pandas; extra == "tests"
Dynamic: license-file

# torchmil

<p align="left">
    <img src="https://img.shields.io/badge/license-Apache_2.0-blue" alt="License">
    <a href="#"><img src="https://badgen.net/github/contributors/Franblueee/torchmil" alt="Contributors"></a>
    <a href="#"><img src="https://badgen.net/github/release/Franblueee/torchmil" alt="Release"></a>
</p>


**torchmil** is a [PyTorch](https://pytorch.org/)-based library for deep Multiple Instance Learning (MIL).
It provides a simple, flexible, and extensible framework for working with MIL models and data.

It includes:

- A collection of popular [MIL models](https://franblueee.github.io/torchmil/api/models/).
- Different [PyTorch modules](https://franblueee.github.io/torchmil/api/nn/) frequently used in MIL models.
- Handy tools to deal with [MIL data](https://franblueee.github.io/torchmil/api/data/).
- A collection of popular [MIL datasets](https://franblueee.github.io/torchmil/api/datasets/).

## Installation

```bash
pip install torchmil
```

## Quick start

You can load a MIL dataset and train a MIL model in just a few lines of code:

```python
from torchmil.datasets import Camelyon16MIL
from torchmil.models import ABMIL
from torchmil.utils import Trainer
from torchmil.data import collate_fn
from torch.utils.data import DataLoader

# Load the Camelyon16 dataset
dataset = Camelyon16MIL(root='data', features='UNI')
dataloader = DataLoader(dataset, batch_size=4, shuffle=True, collate_fn=collate_fn)

# Instantiate the ABMIL model and optimizer
model = ABMIL(in_shape=(2048,), criterion=torch.nn.BCEWithLogitsLoss()) # each model has its own criterion
optimizer = torch.optim.Adam(model.parameters(), lr=1e-4)

# Instantiate the Trainer
trainer = Trainer(model, optimizer, device='cuda')

# Train the model
trainer.train(dataloader, epochs=10)

# Save the model
torch.save(model.state_dict(), 'model.pth')
```

## Next steps

You can take a look at the [examples](https://franblueee.github.io/torchmil/examples/) to see how to use **torchmil** in practice.
To see the full list of available models, datasets, and modules, check the [API reference](https://franblueee.github.io/torchmil/api/).

## Citation

If you find this library useful, please consider citing it:

```bibtex
@misc{torchmil,
  author = {Castro-Mac{\'\i}as, Francisco M and S{\'a}ez-Maldonado, Francisco Javier and Morales Alvarez, Pablo and Molina, Rafael},
  title = {torchmil: A PyTorch-based library for deep Multiple Instance Learning},
  year = {2025},
  howpublished = {\url{https://franblueee.github.io/torchmil/}}
}
```
