<div style="position: relative; width: 100%;">
<img src="docs/images/mainlogo.png" style="width: 250px;">
<a href="https://ccbs.carney.brown.edu/brainstorm" style="position: absolute; right: 0; top: 50%; transform: translateY(-50%);">
<img src="docs/images/Brain-Bolt-%2B-Circuits.gif" style="width: 100px;">
</a>
</div>
## HSSM - Hierarchical Sequential Sampling Modeling
[](https://github.com/ambv/black)
### Overview
HSSM is a Python toolbox that provides a seamless combination of state-of-the-art likelihood approximation methods with the wider ecosystem of probabilistic programming languages. It facilitates flexible hierarchical model building and inference via modern MCMC samplers. HSSM is user-friendly and provides the ability to rigorously estimate the impact of neural and other trial-by-trial covariates through parameter-wise mixed-effects models for a large variety of cognitive process models. HSSM is a <a href="https://ccbs.carney.brown.edu/brainstorm">BRAINSTORM</a> project in collaboration with the Center for Computation and Visualization and the Center for Computational Brain Science within the Carney Institute at Brown University.
- Allows approximate hierarchical Bayesian inference via various likelihood approximators.
- Estimate impact of neural and other trial-by-trial covariates via native hierarchical mixed-regression support.
- Extensible for users to add novel models with corresponding likelihoods.
- Built on PyMC with support from the Python Bayesian ecosystem at large.
- Incorporates Bambi's intuitive `lmer`-like regression parameter specification for within- and between-subject effects.
- Native ArviZ support for plotting and other convenience functions to aid the Bayesian workflow.
- Utilizes the ONNX format for translation of differentiable likelihood approximators across backends.
### [Official documentation](https://lnccbrown.github.io/HSSM/).
## Cite HSSM
Fengler, A., Xu, P., Bera, K., Omar, A., Frank, M.J. (in preparation). HSSM: A generalized toolbox for hierarchical bayesian estimation of computational models in cognitive neuroscience.
## Example
Here is a simple example of how to use HSSM:
```python
import hssm
# Load a package-supplied dataset
cav_data = hssm.load_data('cavanagh_theta')
# Define a basic hierarchical model with trial-level covariates
model = hssm.HSSM(
model="ddm",
data=cav_data,
include=[
{
"name": "v",
"prior": {
"Intercept": {"name": "Normal", "mu": 0.0, "sigma": 0.1},
"theta": {"name": "Normal", "mu": 0.0, "sigma": 0.1},
},
"formula": "v ~ theta + (1|participant_id)",
"link": "identity",
},
],
)
# Sample from the posterior for this model
model.sample()
```
To quickly get started with HSSM, please follow [this tutorial](https://lnccbrown.github.io/HSSM/getting_started/getting_started/).
For a deeper dive into HSSM, please follow [our main tutorial](https://lnccbrown.github.io/HSSM/tutorials/main_tutorial/).
## Installation
HSSM can be directly installed into your conda environment on Linux and MacOS. Installing HSSM on windows takes only one more simple step. We have a more detailed [installation guide](https://lnccbrown.github.io/HSSM/getting_started/installation/) for users with more specific setups.
**Important Update:** From HSSM 0.2.2, the official recommended way to install HSSM is through conda.
### Install HSSM on Linux and MacOS (CPU only)
Use the following command to install HSSM into your virtual environment:
```bash
conda install -c conda-forge hssm
```
### Install HSSM on Linux and MacOS (with GPU Support)
If you need to sample with GPU, please install JAX with GPU support before installing HSSM:
```bash
conda install jaxlib=*=*cuda* jax cuda-nvcc -c conda-forge -c nvidia
conda install -c conda-forge hssm
```
### Install HSSM on Windows (CPU only)
Because `jaxlib` is not available through `conda-forge` on Windows, you need to install JAX on Windows through `pip` before getting HSSM:
```bash
pip install jax
conda install -c conda-forge hssm
```
### Install HSSM on Windows (with GPU support)
You simply need to install JAX with GPU support before getting HSSM:
```bash
pip install jax[cuda12]
conda install -c conda-forge hssm
```
### Support for Apple Silicon, AMD, and other GPUs
JAX also has support other GPUs. Please follow the [Official JAX installation guide](https://jax.readthedocs.io/en/latest/installation.html) to install the correct version of JAX before installing HSSM.
## Advanced Installation
### Install HSSM directly with Pip
HSSM is also available through PyPI. You can directly install it with pip into any virtual environment via:
```bash
pip install hssm
```
**Note:** While this installation is much simpler, you might encounter this warning message `WARNING (pytensor.tensor.blas): Using NumPy C-API based implementation for BLAS functions.` Please refer to our [advanced installation guide](https://lnccbrown.github.io/HSSM/getting_started/installation/) for more details.
### Install the dev version of HSSM
You can install the dev version of `hssm` directly from this repo:
```bash
pip install git+https://github.com/lnccbrown/HSSM.git
```
### Install HSSM on Google Colab
Google Colab comes with PyMC and JAX pre-configured. That holds true even if you are using the GPU and TPU backend, so you simply need to install HSSM via pip on Colab regardless of the backend you are using:
```bash
!pip install hssm
```
## Troubleshooting
**Note:** Possible solutions to any issues with installations with hssm can be located
[here](https://github.com/lnccbrown/HSSM/discussions). Also feel free to start a new
discussion thread if you don't find answers there. We recommend installing HSSM into
a new conda environment with Python 3.10 or 3.11 to prevent any problems with dependencies
during the installation process. Please note that hssm is only tested for python 3.10,
3.11. As of HSSM v0.2.0, support for Python 3.9 is dropped. Use unsupported python
versions with caution.
## License
HSSM is licensed under [Copyright 2023, Brown University, Providence, RI](LICENSE)
## Support
For questions, please feel free to [open a discussion](https://github.com/lnccbrown/HSSM/discussions).
For bug reports and feature requests, please feel free to [open an issue](https://github.com/lnccbrown/HSSM/issues) using the corresponding template.
## Contribution
If you want to contribute to this project, please follow our [contribution guidelines](docs/CONTRIBUTING.md).
## Acknowledgements
We would like to extend our gratitude to the following individuals for their valuable contributions to the development of the HSSM package:
- [Bambi](https://github.com/bambinos/bambi) - A special thanks to the Bambi project for providing inspiration, guidance, and support throughout the development process. [Tomás Capretto](https://github.com/tomicapretto), a key contributor to Bambi, provided invaluable assistance in the development of the HSSM package.
Those contributions have greatly enhanced the functionality and quality of the HSSM.
Raw data
{
"_id": null,
"home_page": "https://github.com/lnccbrown/HSSM",
"name": "HSSM",
"maintainer": null,
"docs_url": null,
"requires_python": "<=3.12,>=3.10",
"maintainer_email": null,
"keywords": "HSSM, sequential sampling models, bayesian, bayes, mcmc",
"author": "Alexander Fengler",
"author_email": "alexander_fengler@brown.edu",
"download_url": "https://files.pythonhosted.org/packages/94/15/6e7db147f64737e25ee232ce4ce04bc181497ba0ac16b1f92e9b9276ed54/hssm-0.2.4.tar.gz",
"platform": null,
"description": "<div style=\"position: relative; width: 100%;\">\n <img src=\"docs/images/mainlogo.png\" style=\"width: 250px;\">\n <a href=\"https://ccbs.carney.brown.edu/brainstorm\" style=\"position: absolute; right: 0; top: 50%; transform: translateY(-50%);\">\n <img src=\"docs/images/Brain-Bolt-%2B-Circuits.gif\" style=\"width: 100px;\">\n </a>\n</div>\n\n## HSSM - Hierarchical Sequential Sampling Modeling\n\n\n\n\n\n\n\n[](https://github.com/ambv/black)\n\n### Overview\n\nHSSM is a Python toolbox that provides a seamless combination of state-of-the-art likelihood approximation methods with the wider ecosystem of probabilistic programming languages. It facilitates flexible hierarchical model building and inference via modern MCMC samplers. HSSM is user-friendly and provides the ability to rigorously estimate the impact of neural and other trial-by-trial covariates through parameter-wise mixed-effects models for a large variety of cognitive process models. HSSM is a <a href=\"https://ccbs.carney.brown.edu/brainstorm\">BRAINSTORM</a> project in collaboration with the Center for Computation and Visualization and the Center for Computational Brain Science within the Carney Institute at Brown University.\n\n- Allows approximate hierarchical Bayesian inference via various likelihood approximators.\n- Estimate impact of neural and other trial-by-trial covariates via native hierarchical mixed-regression support.\n- Extensible for users to add novel models with corresponding likelihoods.\n- Built on PyMC with support from the Python Bayesian ecosystem at large.\n- Incorporates Bambi's intuitive `lmer`-like regression parameter specification for within- and between-subject effects.\n- Native ArviZ support for plotting and other convenience functions to aid the Bayesian workflow.\n- Utilizes the ONNX format for translation of differentiable likelihood approximators across backends.\n\n### [Official documentation](https://lnccbrown.github.io/HSSM/).\n\n## Cite HSSM\n\nFengler, A., Xu, P., Bera, K., Omar, A., Frank, M.J. (in preparation). HSSM: A generalized toolbox for hierarchical bayesian estimation of computational models in cognitive neuroscience.\n\n## Example\n\nHere is a simple example of how to use HSSM:\n\n```python\nimport hssm\n\n# Load a package-supplied dataset\ncav_data = hssm.load_data('cavanagh_theta')\n\n# Define a basic hierarchical model with trial-level covariates\nmodel = hssm.HSSM(\n model=\"ddm\",\n data=cav_data,\n include=[\n {\n \"name\": \"v\",\n \"prior\": {\n \"Intercept\": {\"name\": \"Normal\", \"mu\": 0.0, \"sigma\": 0.1},\n \"theta\": {\"name\": \"Normal\", \"mu\": 0.0, \"sigma\": 0.1},\n },\n \"formula\": \"v ~ theta + (1|participant_id)\",\n \"link\": \"identity\",\n },\n ],\n)\n\n# Sample from the posterior for this model\nmodel.sample()\n```\n\nTo quickly get started with HSSM, please follow [this tutorial](https://lnccbrown.github.io/HSSM/getting_started/getting_started/).\nFor a deeper dive into HSSM, please follow [our main tutorial](https://lnccbrown.github.io/HSSM/tutorials/main_tutorial/).\n\n## Installation\n\nHSSM can be directly installed into your conda environment on Linux and MacOS. Installing HSSM on windows takes only one more simple step. We have a more detailed [installation guide](https://lnccbrown.github.io/HSSM/getting_started/installation/) for users with more specific setups.\n\n**Important Update:** From HSSM 0.2.2, the official recommended way to install HSSM is through conda.\n\n### Install HSSM on Linux and MacOS (CPU only)\n\nUse the following command to install HSSM into your virtual environment:\n\n```bash\nconda install -c conda-forge hssm\n```\n\n### Install HSSM on Linux and MacOS (with GPU Support)\n\nIf you need to sample with GPU, please install JAX with GPU support before installing HSSM:\n\n```bash\nconda install jaxlib=*=*cuda* jax cuda-nvcc -c conda-forge -c nvidia\nconda install -c conda-forge hssm\n```\n\n### Install HSSM on Windows (CPU only)\n\nBecause `jaxlib` is not available through `conda-forge` on Windows, you need to install JAX on Windows through `pip` before getting HSSM:\n\n```bash\npip install jax\nconda install -c conda-forge hssm\n```\n\n### Install HSSM on Windows (with GPU support)\n\nYou simply need to install JAX with GPU support before getting HSSM:\n\n```bash\npip install jax[cuda12]\nconda install -c conda-forge hssm\n```\n\n### Support for Apple Silicon, AMD, and other GPUs\n\nJAX also has support other GPUs. Please follow the [Official JAX installation guide](https://jax.readthedocs.io/en/latest/installation.html) to install the correct version of JAX before installing HSSM.\n\n\n## Advanced Installation\n\n### Install HSSM directly with Pip\n\nHSSM is also available through PyPI. You can directly install it with pip into any virtual environment via:\n\n```bash\npip install hssm\n```\n\n**Note:** While this installation is much simpler, you might encounter this warning message `WARNING (pytensor.tensor.blas): Using NumPy C-API based implementation for BLAS functions.` Please refer to our [advanced installation guide](https://lnccbrown.github.io/HSSM/getting_started/installation/) for more details.\n\n### Install the dev version of HSSM\n\nYou can install the dev version of `hssm` directly from this repo:\n\n```bash\npip install git+https://github.com/lnccbrown/HSSM.git\n```\n\n### Install HSSM on Google Colab\n\nGoogle Colab comes with PyMC and JAX pre-configured. That holds true even if you are using the GPU and TPU backend, so you simply need to install HSSM via pip on Colab regardless of the backend you are using:\n\n```bash\n!pip install hssm\n```\n\n## Troubleshooting\n\n**Note:** Possible solutions to any issues with installations with hssm can be located\n[here](https://github.com/lnccbrown/HSSM/discussions). Also feel free to start a new\ndiscussion thread if you don't find answers there. We recommend installing HSSM into\na new conda environment with Python 3.10 or 3.11 to prevent any problems with dependencies\nduring the installation process. Please note that hssm is only tested for python 3.10,\n3.11. As of HSSM v0.2.0, support for Python 3.9 is dropped. Use unsupported python\nversions with caution.\n\n## License\n\nHSSM is licensed under [Copyright 2023, Brown University, Providence, RI](LICENSE)\n\n## Support\n\nFor questions, please feel free to [open a discussion](https://github.com/lnccbrown/HSSM/discussions).\n\nFor bug reports and feature requests, please feel free to [open an issue](https://github.com/lnccbrown/HSSM/issues) using the corresponding template.\n\n## Contribution\n\nIf you want to contribute to this project, please follow our [contribution guidelines](docs/CONTRIBUTING.md).\n\n## Acknowledgements\n\nWe would like to extend our gratitude to the following individuals for their valuable contributions to the development of the HSSM package:\n\n- [Bambi](https://github.com/bambinos/bambi) - A special thanks to the Bambi project for providing inspiration, guidance, and support throughout the development process. [Tom\u00e1s Capretto](https://github.com/tomicapretto), a key contributor to Bambi, provided invaluable assistance in the development of the HSSM package.\n\nThose contributions have greatly enhanced the functionality and quality of the HSSM.\n",
"bugtrack_url": null,
"license": "Copyright 2023, Brown University, Providence, RI.",
"summary": "Bayesian inference for hierarchical sequential sampling models.",
"version": "0.2.4",
"project_urls": {
"Homepage": "https://github.com/lnccbrown/HSSM",
"Repository": "https://github.com/lnccbrown/HSSM"
},
"split_keywords": [
"hssm",
" sequential sampling models",
" bayesian",
" bayes",
" mcmc"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8a2382ca493371700c7d2ff37380b65fd44d59508d769e8b5a62a52168929d9e",
"md5": "ecd15faf993d4876e81613fbd9b8037d",
"sha256": "0017c394fb32ae85162c9df416d30a17d4cfada6b65903798ef5fe1a527fe7ab"
},
"downloads": -1,
"filename": "hssm-0.2.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "ecd15faf993d4876e81613fbd9b8037d",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<=3.12,>=3.10",
"size": 181222,
"upload_time": "2024-09-18T23:30:59",
"upload_time_iso_8601": "2024-09-18T23:30:59.172001Z",
"url": "https://files.pythonhosted.org/packages/8a/23/82ca493371700c7d2ff37380b65fd44d59508d769e8b5a62a52168929d9e/hssm-0.2.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "94156e7db147f64737e25ee232ce4ce04bc181497ba0ac16b1f92e9b9276ed54",
"md5": "d29d32e0bb1ba09015f5a2fa9f347057",
"sha256": "5bd1986da145942cf619e4cd7ccb0f00d59f20cc05780acdf531800e98c19dd1"
},
"downloads": -1,
"filename": "hssm-0.2.4.tar.gz",
"has_sig": false,
"md5_digest": "d29d32e0bb1ba09015f5a2fa9f347057",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<=3.12,>=3.10",
"size": 172437,
"upload_time": "2024-09-18T23:31:00",
"upload_time_iso_8601": "2024-09-18T23:31:00.634963Z",
"url": "https://files.pythonhosted.org/packages/94/15/6e7db147f64737e25ee232ce4ce04bc181497ba0ac16b1f92e9b9276ed54/hssm-0.2.4.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-09-18 23:31:00",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "lnccbrown",
"github_project": "HSSM",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "hssm"
}
JSON
