.github/README.md

<div align="center">
    <h1>MLIP Arena</h1>
    <a href="https://github.com/atomind-ai/mlip-arena/actions"><img alt="GitHub Actions Workflow Status" src="https://img.shields.io/github/actions/workflow/status/atomind-ai/mlip-arena/test.yaml"></a>
    <a href="https://pypi.org/project/mlip-arena/"><img alt="PyPI - Version" src="https://img.shields.io/pypi/v/mlip-arena"></a>
    <a href="https://zenodo.org/doi/10.5281/zenodo.13704399"><img src="https://zenodo.org/badge/776930320.svg" alt="DOI"></a>
    <a href="https://huggingface.co/spaces/atomind/mlip-arena"><img src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Space-blue" alt="Hugging Face"></a>
    <!-- <a href="https://discord.gg/W8WvdQtT8T"><img alt="Discord" src="https://img.shields.io/discord/1299613474820984832?logo=discord"> -->
</a>
</div>

> [!CAUTION]
> MLIP Arena is currently in pre-alpha. The results are not stable. Please intepret them with care. 

> [!NOTE]
> Contributions of new tasks are very welcome! If you're interested in joining the effort, please reach out to Yuan at [[email protected]](mailto:[email protected]). See [project page](https://github.com/orgs/atomind-ai/projects/1) for some outstanding tasks, or propose new one in [Discussion](https://github.com/atomind-ai/mlip-arena/discussions/new?category=ideas).

MLIP Arena is a unified platform for evaluating foundation machine learning interatomic potentials (MLIPs) beyond conventional error metrics. It focuses on revealing the physics and chemistry learned by these models and assessing their utilitarian performance agnostic to underlying model architecture. The platform's benchmarks are specifically designed to evaluate the readiness and reliability of open-source, open-weight models in accurately reproducing both qualitative and quantitative behaviors of atomic systems.

MLIP Arena leverages modern pythonic workflow orchestrator [Prefect](https://www.prefect.io/) to enable advanced task/flow chaining and caching.

## Installation

### From PyPI (without model running capability)

```bash
pip install mlip-arena
```

### From source

**Linux**

```bash
# (Optional) Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
source $HOME/.local/bin/env
# One script uv pip installation
bash scripts/install-linux.sh
```

```bash
# Or from command line
git clone https://github.com/atomind-ai/mlip-arena.git
cd mlip-arena
pip install torch==2.2.0
bash scripts/install-pyg.sh
bash scripts/install-dgl.sh
pip install -e .[test]
pip install -e .[mace]
# DeePMD
DP_ENABLE_TENSORFLOW=0 pip install -e .[deepmd]
```

**Mac**

```bash
# (Optional) Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
source $HOME/.local/bin/env
# One script uv pip installation
bash scripts/install-macosx.sh
```

## Quickstart

### Molecular dynamics (MD)

Arena provides a unified interface to run all the compiled MLIPs. This can be achieved simply by looping through `MLIPEnum`:

```python
from mlip_arena.models import MLIPEnum
from mlip_arena.tasks.md import run as MD 
# from mlip_arena.tasks import MD # for convenient import
from mlip_arena.tasks.utils import get_calculator

from ase import units
from ase.build import bulk

atoms = bulk("Cu", "fcc", a=3.6)

results = []

for model in MLIPEnum:
    result = MD(
        atoms=atoms,
        calculator=get_calculator(
            model,
            calculator_kwargs=dict(), # passing into calculator
            dispersion=True,
            dispersion_kwargs=dict(damping='bj', xc='pbe', cutoff=40.0 * units.Bohr), # passing into TorchDFTD3Calculator
        ),
        ensemble="nve",
        dynamics="velocityverlet",
        total_time=1e3, # 1 ps = 1e3 fs
        time_step=2, # fs
    )
    results.append(result)
```

### List of implemented tasks

The implemented tasks are available under `mlip_arena.tasks.<module>.run` or `from mlip_arena.tasks import *` for convenient imports (currently doesn't work if [phonopy](https://phonopy.github.io/phonopy/install.html) is not installed).

- [OPT](../mlip_arena/tasks/optimize.py#L56): Structure optimization
- [EOS](../mlip_arena/tasks/eos.py#L42): Equation of state (energy-volume scan)
- [MD](../mlip_arena/tasks/md.py#L200): Molecular dynamics with flexible dynamics (NVE, NVT, NPT) and temperature/pressure scheduling (annealing, shearing, *etc*)
- [PHONON](../mlip_arena/tasks/phonon.py#L110): Phonon calculation driven by [phonopy](https://phonopy.github.io/phonopy/install.html)
- [NEB](../mlip_arena/tasks/neb.py#L96): Nudged elastic band
- [NEB_FROM_ENDPOINTS](../mlip_arena/tasks/neb.py#L164): Nudge elastic band with convenient image interpolation (linear or IDPP)
- [ELASTICITY](../mlip_arena/tasks/elasticity.py#L78): Elastic tensor calculation


## Contribute

MLIP Arena is now in pre-alpha. If you're interested in joining the effort, please reach out to Yuan at [[email protected]](mailto:[email protected]). 

### Development

```
git lfs fetch --all
git lfs pull
streamlit run serve/app.py
```

### Add new benchmark tasks (WIP)

> [!NOTE]
> Please reuse, extend, or chain the general tasks defined [above](#list-of-implemented-tasks)

<!-- 1. Follow the task template to implement the task class and upload the script along with metadata to the MLIP Arena [here](../mlip_arena/tasks/README.md).
1. Code a benchmark script to evaluate the performance of your model on the task. The script should be able to load the model and the dataset, and output the evaluation metrics. -->

### Add new MLIP models 

If you have pretrained MLIP models that you would like to contribute to the MLIP Arena and show benchmark in real-time, there are two ways:

#### External ASE Calculator (easy)

1. Implement new ASE Calculator class in [mlip_arena/models/externals](../mlip_arena/models/externals). 
2. Name your class with awesome model name and add the same name to [registry](../mlip_arena/models/registry.yaml) with metadata.

> [!CAUTION] 
> Remove unneccessary outputs under `results` class attributes to avoid error for MD simulations. Please refer to other class definition for example.

#### Hugging Face Model (recommended, difficult)

0. Inherit Hugging Face [ModelHubMixin](https://huggingface.co/docs/huggingface_hub/en/package_reference/mixins) class to your awesome model class definition. We recommend [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/en/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin).
1. Create a new [Hugging Face Model](https://huggingface.co/new) repository and upload the model file using [push_to_hub function](https://huggingface.co/docs/huggingface_hub/en/package_reference/mixins#huggingface_hub.ModelHubMixin.push_to_hub).
2. Follow the template to code the I/O interface for your model [here](../mlip_arena/models/README.md). 
3. Update model [registry](../mlip_arena/models/registry.yaml) with metadata

<!-- > [!NOTE] 
> CPU benchmarking will be performed automatically. Due to the limited amount GPU compute, if you would like to be considered for GPU benchmarking, please create a pull request to demonstrate the offline performance of your model (published paper or preprint). We will review and select the models to be benchmarked on GPU. -->

<!-- ### Add new datasets

The "ultimate" goal is to compile the copies of all the open data in a unified format for lifelong learning with [Hugging Face Auto-Train](https://huggingface.co/docs/hub/webhooks-guide-auto-retrain). 

1. Create a new [Hugging Face Dataset](https://huggingface.co/new-dataset) repository and upload the reference data (e.g. DFT, AIMD, experimental measurements such as RDF).

#### Single-point density functional theory calculations

- [ ] MPTrj
- [ ] [Alexandria](https://huggingface.co/datasets/atomind/alexandria)
- [ ] QM9
- [ ] SPICE

#### Molecular dynamics calculations

- [ ] [MD17](http://www.sgdml.org/#datasets)
- [ ] [MD22](http://www.sgdml.org/#datasets) -->