# lintsampler
**Efficient random sampling via linear interpolation.**
[](https://github.com/aneeshnaik/lintsampler/actions/workflows/test.yaml)
[](https://coveralls.io/github/aneeshnaik/lintsampler?branch=main)
[](https://lintsampler.readthedocs.io/en/latest/?badge=latest)
[](https://github.com/aneeshnaik/lintsampler/blob/main/LICENSE)
When you have a density function, but you would like to create a set of sample points from that density function, you can use _linear interpolate sampling_. Using the evaluation of the density at the two endpoints of 1D interval, or the four corners of a 2D rectangle, or generally the $2^k$ vertices of a $k$-dimensional hyperbox (or a series of such hyperboxes, e.g., the cells of a $k$-dimensional grid), linear interpolant sampling is a technique to draw random samples within the hyperbox. `lintsampler` provides a Python implementation of this.
See the [documentation](https://lintsampler.readthedocs.io/) or our [paper](https://github.com/aneeshnaik/lintsampler/blob/main/paper/paper.pdf) for further details.
## Installation
Three ways of installing `lintsampler`:
- `pip`:
```
pip install lintsampler
```
- `conda`:
```
conda install -c conda-forge lintsampler
```
- Simply cloning this repository.
## Quickstart Example
If you have a density function, such as this multi-modal 1d pdf with the bulk of the density between -7 and 7,
```python
import numpy as np
def gmm_pdf(x):
mu = np.array([-3.0, 0.5, 2.5])
sig = np.array([1.0, 0.25, 0.75])
w = np.array([0.4, 0.25, 0.35])
return np.sum([w[i] * norm.pdf(x, mu[i], sig[i]) for i in range(3)], axis=0)
```
`lintsampler` can efficiently draw samples from it on some defined interval (here a 100-point grid between -7 and 7):
```python
from lintsampler import LintSampler
grid = np.linspace(-7,7,100)
samples = LintSampler(grid,pdf=gmm_pdf).sample(N=10000)
```
Making a histogram of the resulting samples and comparing to the input density function shows good agreement -- and we can do even better by increasing the resolution.
See [this page of the documentation](https://lintsampler.readthedocs.io/en/latest/examples/1_gmm.html) for a more detailed explanation of this example.
## Documentation
Complete documentation, including more example notebooks, is available at [lintsampler.readthedocs.io/](https://lintsampler.readthedocs.io/).
## Attribution
If using `lintsampler` for a research publication, please cite our [paper](https://github.com/aneeshnaik/lintsampler/blob/main/paper/paper.pdf).
## License
`lintsampler` is available under the MIT license. See the LICENSE file for specifics.
Raw data
{
"_id": null,
"home_page": null,
"name": "lintsampler",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "python, statistics, numpy, sampling, random",
"author": null,
"author_email": "\"Aneesh P. Naik\" <aneesh.naik@roe.ac.uk>, \"Michael S. Petersen\" <michael.petersen@roe.ac.uk>",
"download_url": "https://files.pythonhosted.org/packages/25/13/3b9e60c45968cc4930917cf6992f8db76492c62915bf913f1b8b928ae94f/lintsampler-0.3.0.tar.gz",
"platform": null,
"description": "# lintsampler\n\n\n**Efficient random sampling via linear interpolation.**\n\n[](https://github.com/aneeshnaik/lintsampler/actions/workflows/test.yaml)\n[](https://coveralls.io/github/aneeshnaik/lintsampler?branch=main)\n[](https://lintsampler.readthedocs.io/en/latest/?badge=latest)\n[](https://github.com/aneeshnaik/lintsampler/blob/main/LICENSE)\n\nWhen you have a density function, but you would like to create a set of sample points from that density function, you can use _linear interpolate sampling_. Using the evaluation of the density at the two endpoints of 1D interval, or the four corners of a 2D rectangle, or generally the $2^k$ vertices of a $k$-dimensional hyperbox (or a series of such hyperboxes, e.g., the cells of a $k$-dimensional grid), linear interpolant sampling is a technique to draw random samples within the hyperbox. `lintsampler` provides a Python implementation of this.\n\nSee the [documentation](https://lintsampler.readthedocs.io/) or our [paper](https://github.com/aneeshnaik/lintsampler/blob/main/paper/paper.pdf) for further details. \n\n## Installation\n\nThree ways of installing `lintsampler`:\n\n- `pip`:\n```\npip install lintsampler\n```\n\n- `conda`:\n```\nconda install -c conda-forge lintsampler\n```\n\n- Simply cloning this repository.\n\n## Quickstart Example\n\nIf you have a density function, such as this multi-modal 1d pdf with the bulk of the density between -7 and 7,\n\n```python\nimport numpy as np\n\ndef gmm_pdf(x):\n mu = np.array([-3.0, 0.5, 2.5])\n sig = np.array([1.0, 0.25, 0.75])\n w = np.array([0.4, 0.25, 0.35])\n return np.sum([w[i] * norm.pdf(x, mu[i], sig[i]) for i in range(3)], axis=0)\n```\n\n`lintsampler` can efficiently draw samples from it on some defined interval (here a 100-point grid between -7 and 7):\n\n```python\nfrom lintsampler import LintSampler\n\ngrid = np.linspace(-7,7,100)\nsamples = LintSampler(grid,pdf=gmm_pdf).sample(N=10000)\n```\n\nMaking a histogram of the resulting samples and comparing to the input density function shows good agreement -- and we can do even better by increasing the resolution.\n\n\nSee [this page of the documentation](https://lintsampler.readthedocs.io/en/latest/examples/1_gmm.html) for a more detailed explanation of this example.\n\n## Documentation\n\nComplete documentation, including more example notebooks, is available at [lintsampler.readthedocs.io/](https://lintsampler.readthedocs.io/).\n\n## Attribution\n\nIf using `lintsampler` for a research publication, please cite our [paper](https://github.com/aneeshnaik/lintsampler/blob/main/paper/paper.pdf).\n\n## License\n\n`lintsampler` is available under the MIT license. See the LICENSE file for specifics.\n",
"bugtrack_url": null,
"license": null,
"summary": "Efficient random sampling via linear interpolation.",
"version": "0.3.0",
"project_urls": {
"Documentation": "https://lintsampler.readthedocs.io",
"Homepage": "https://github.com/aneeshnaik/lintsampler"
},
"split_keywords": [
"python",
" statistics",
" numpy",
" sampling",
" random"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "734ba1aea1d52948bed44718b82775fc01abf76112f84b438246029f787c05c6",
"md5": "c575d8a5b995056eb60a11ad4056e653",
"sha256": "db6fd54a327b135a1f8f932fca2d498c7c6ac5f484ff2ca7bb9cfdb07b549e1f"
},
"downloads": -1,
"filename": "lintsampler-0.3.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "c575d8a5b995056eb60a11ad4056e653",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 30467,
"upload_time": "2024-06-14T10:22:09",
"upload_time_iso_8601": "2024-06-14T10:22:09.163079Z",
"url": "https://files.pythonhosted.org/packages/73/4b/a1aea1d52948bed44718b82775fc01abf76112f84b438246029f787c05c6/lintsampler-0.3.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "25133b9e60c45968cc4930917cf6992f8db76492c62915bf913f1b8b928ae94f",
"md5": "9674dafa85aadabf4475d95a97b2fd4c",
"sha256": "dc8ab38e183bac59988a4be2b479385fae1fff55904aeb54024c27bcfc24e532"
},
"downloads": -1,
"filename": "lintsampler-0.3.0.tar.gz",
"has_sig": false,
"md5_digest": "9674dafa85aadabf4475d95a97b2fd4c",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 13676309,
"upload_time": "2024-06-14T10:22:11",
"upload_time_iso_8601": "2024-06-14T10:22:11.858681Z",
"url": "https://files.pythonhosted.org/packages/25/13/3b9e60c45968cc4930917cf6992f8db76492c62915bf913f1b8b928ae94f/lintsampler-0.3.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-06-14 10:22:11",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "aneeshnaik",
"github_project": "lintsampler",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"test_requirements": [],
"lcname": "lintsampler"
}
JSON
