# elapid
<img src="https://earth-chris.github.io/elapid/img/amazon.jpg" alt="the amazon"/>
<p align="center">
<em>Contemporary species distribution modeling tools for python.</em>
</p>
[](https://earth-chris.github.io/elapid/paper/draft-paper.pdf)
---
**Documentation**: [earth-chris.github.io/elapid](https://earth-chris.github.io/elapid)
**Source code**: [earth-chris/elapid](https://github.com/earth-chris/elapid)
---
## :snake: Introduction
`elapid` is a series of species distribution modeling tools for python. This includes a custom implementation of [Maxent][home-maxent] and a suite of methods to simplify working with biogeography data.
The name is an homage to *A Biogeographic Analysis of Australian Elapid Snakes* (H.A. Nix, 1986), the paper widely credited with defining the essential bioclimatic variables to use in species distribution modeling. It's also a snake pun (a python wrapper for mapping snake biogeography).
---
## :seedling: Installation
`pip install elapid` or `conda install -c conda-forge elapid`
Installing `glmnet` is optional, but recommended. This can be done with `pip install elapid[glmnet]` or `conda install -c conda-forge elapid glmnet`. For more support, and for information on why this package is recommended, see [this page](https://elapid.org/install#installing-glmnet).
The `conda` install is recommended for Windows users. While there is a `pip` distribution, you may experience some challenges. The easiest way to overcome them is to use [Windows Subsystem for Linux (WSL)](https://docs.microsoft.com/en-us/windows/wsl/about). Otherwise, see [this page](https://elapid.org/install) for support.
---
## :deciduous_tree: Why use elapid?
The amount and quality of bioegeographic data has increased dramatically over the past decade, as have cloud-based tools for working with it. `elapid` was designed to provide a set of modern, python-based tools for working with species occurrence records and environmental covariates to map different dimensions of a species' niche.
`elapid` supports working with modern geospatial data formats and uses contemporary approaches to training statistical models. It uses `sklearn` conventions to fit and apply models, `rasterio` to handle raster operations, `geopandas` for vector operations, and processes data under the hood with `numpy`.
This makes it easier to do things like fit/apply models to multi-temporal and multi-scale data, fit geographically-weighted models, create ensembles, precisely define background point distributions, and summarize model predictions.
It does the following things reasonably well:
:globe_with_meridians: **Point sampling**
Select random geographic point samples (aka background or pseudoabsence points) within polygons or rasters, handling `nodata` locations, as well as sampling from bias maps (using `elapid.sample_raster()`, `elapid.sample_vector()`, or `elapid.sample_bias_file()`).
:chart_with_upwards_trend: **Vector annotation**
Extract and annotate point data from rasters, creating `GeoDataFrames` with sample locations and their matching covariate values (using `elapid.annotate()`). On-the-fly reprojection, dropping nodata, multi-band inputs and multi-file inputs are all supported.
:bar_chart: **Zonal statistics**
Calculate zonal statistics from multi-band, multi-raster data into a single `GeoDataFrame` from one command (using `elapid.zonal_stats()`).
:bug: **Feature transformations**
Transform covariate data into derivative `features` to expand data dimensionality and improve prediction accuracy (like `elapid.ProductTransformer()`, `elapid.HingeTransformer()`, or the all-in-one `elapid.MaxentFeatureTransformer()`).
:bird: **Species distribution modeling**
Train and apply species distribution models based on annotated point data, configured with sensible defaults (like `elapid.MaxentModel()` and `elapid.NicheEnvelopeModel()`).
:satellite: **Training spatially-aware models**
Compute spatially-explicit sample weights, checkerboard train/test splits, or geographically-clustered cross-validation splits to reduce spatial autocorellation effects (with `elapid.distance_weights()`, `elapid.checkerboard_split()` and `elapid.GeographicKFold()`).
:earth_asia: **Applying models to rasters**
Apply any pixel-based model with a `.predict()` method to raster data to easily create prediction probability maps (like training a `RandomForestClassifier()` and applying with `elapid.apply_model_to_rasters()`).
:cloud: **Cloud-native geo support**
Work with cloud- or web-hosted raster/vector data (on `https://`, `gs://`, `s3://`, etc.) to keep your disk free of temporary files.
Check out some example code snippets and workflows on the [Working with Geospatial Data](https://elapid.org/examples/WorkingWithGeospatialData/) page.
---
:snake: `elapid` requires some effort on the user's part to draw samples and extract covariate data. This is by design.
Selecting background samples, computing sample weights, splitting train/test data, and specifying training parameters are all critical modeling choices that have profound effects on inference and interpretation.
The extra flexibility provided by `elapid` enables more control over the seemingly black-box approach of Maxent, enabling users to better tune and evaluate their models.
---
## Developed by
[Christopher Anderson](https://cbanderson.info)[^1] [^2]
<a href="https://twitter.com/earth_chris"></a>
<a href="https://github.com/earth-chris"></a>
[home-maxent]: https://biodiversityinformatics.amnh.org/open_source/maxent/
[r-maxnet]: https://github.com/mrmaxent/maxnet
[^1]: [EO Lab, Planet Labs PBC](https://www.planet.com)
[^2]: [Center for Conservation Biology, Stanford University](https://ccb.stanford.edu)
Raw data
{
"_id": null,
"home_page": "https://elapid.org",
"name": "elapid",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.7.0",
"maintainer_email": "",
"keywords": "biogeography,ecology,conservation,SDM,species distribution modeling,maxent",
"author": "Christopher Anderson",
"author_email": "cbanders@stanford.edu",
"download_url": "https://files.pythonhosted.org/packages/fe/8d/2e1ee247b1c8016a58f4bd3bcad910f240a7b03837fa04c6e2348acfcd80/elapid-1.0.1.tar.gz",
"platform": "any",
"description": "# elapid\n\n<img src=\"https://earth-chris.github.io/elapid/img/amazon.jpg\" alt=\"the amazon\"/>\n\n<p align=\"center\">\n <em>Contemporary species distribution modeling tools for python.</em>\n</p>\n\n\n\n\n\n\n[](https://earth-chris.github.io/elapid/paper/draft-paper.pdf)\n\n---\n\n**Documentation**: [earth-chris.github.io/elapid](https://earth-chris.github.io/elapid)\n\n**Source code**: [earth-chris/elapid](https://github.com/earth-chris/elapid)\n\n---\n\n## :snake: Introduction\n\n`elapid` is a series of species distribution modeling tools for python. This includes a custom implementation of [Maxent][home-maxent] and a suite of methods to simplify working with biogeography data.\n\nThe name is an homage to *A Biogeographic Analysis of Australian Elapid Snakes* (H.A. Nix, 1986), the paper widely credited with defining the essential bioclimatic variables to use in species distribution modeling. It's also a snake pun (a python wrapper for mapping snake biogeography).\n\n---\n\n## :seedling: Installation\n\n`pip install elapid` or `conda install -c conda-forge elapid`\n\nInstalling `glmnet` is optional, but recommended. This can be done with `pip install elapid[glmnet]` or `conda install -c conda-forge elapid glmnet`. For more support, and for information on why this package is recommended, see [this page](https://elapid.org/install#installing-glmnet).\n\nThe `conda` install is recommended for Windows users. While there is a `pip` distribution, you may experience some challenges. The easiest way to overcome them is to use [Windows Subsystem for Linux (WSL)](https://docs.microsoft.com/en-us/windows/wsl/about). Otherwise, see [this page](https://elapid.org/install) for support.\n\n---\n\n## :deciduous_tree: Why use elapid?\n\nThe amount and quality of bioegeographic data has increased dramatically over the past decade, as have cloud-based tools for working with it. `elapid` was designed to provide a set of modern, python-based tools for working with species occurrence records and environmental covariates to map different dimensions of a species' niche.\n\n`elapid` supports working with modern geospatial data formats and uses contemporary approaches to training statistical models. It uses `sklearn` conventions to fit and apply models, `rasterio` to handle raster operations, `geopandas` for vector operations, and processes data under the hood with `numpy`.\n\nThis makes it easier to do things like fit/apply models to multi-temporal and multi-scale data, fit geographically-weighted models, create ensembles, precisely define background point distributions, and summarize model predictions.\n\nIt does the following things reasonably well:\n\n:globe_with_meridians: **Point sampling**\n\nSelect random geographic point samples (aka background or pseudoabsence points) within polygons or rasters, handling `nodata` locations, as well as sampling from bias maps (using `elapid.sample_raster()`, `elapid.sample_vector()`, or `elapid.sample_bias_file()`).\n\n:chart_with_upwards_trend: **Vector annotation**\n\nExtract and annotate point data from rasters, creating `GeoDataFrames` with sample locations and their matching covariate values (using `elapid.annotate()`). On-the-fly reprojection, dropping nodata, multi-band inputs and multi-file inputs are all supported.\n\n:bar_chart: **Zonal statistics**\n\nCalculate zonal statistics from multi-band, multi-raster data into a single `GeoDataFrame` from one command (using `elapid.zonal_stats()`).\n\n:bug: **Feature transformations**\n\nTransform covariate data into derivative `features` to expand data dimensionality and improve prediction accuracy (like `elapid.ProductTransformer()`, `elapid.HingeTransformer()`, or the all-in-one `elapid.MaxentFeatureTransformer()`).\n\n:bird: **Species distribution modeling**\n\nTrain and apply species distribution models based on annotated point data, configured with sensible defaults (like `elapid.MaxentModel()` and `elapid.NicheEnvelopeModel()`).\n\n:satellite: **Training spatially-aware models**\n\nCompute spatially-explicit sample weights, checkerboard train/test splits, or geographically-clustered cross-validation splits to reduce spatial autocorellation effects (with `elapid.distance_weights()`, `elapid.checkerboard_split()` and `elapid.GeographicKFold()`).\n\n:earth_asia: **Applying models to rasters**\n\nApply any pixel-based model with a `.predict()` method to raster data to easily create prediction probability maps (like training a `RandomForestClassifier()` and applying with `elapid.apply_model_to_rasters()`).\n\n:cloud: **Cloud-native geo support**\n\nWork with cloud- or web-hosted raster/vector data (on `https://`, `gs://`, `s3://`, etc.) to keep your disk free of temporary files.\n\nCheck out some example code snippets and workflows on the [Working with Geospatial Data](https://elapid.org/examples/WorkingWithGeospatialData/) page.\n\n---\n\n:snake: `elapid` requires some effort on the user's part to draw samples and extract covariate data. This is by design.\n\nSelecting background samples, computing sample weights, splitting train/test data, and specifying training parameters are all critical modeling choices that have profound effects on inference and interpretation.\n\nThe extra flexibility provided by `elapid` enables more control over the seemingly black-box approach of Maxent, enabling users to better tune and evaluate their models.\n\n---\n\n## Developed by\n\n[Christopher Anderson](https://cbanderson.info)[^1] [^2]\n\n<a href=\"https://twitter.com/earth_chris\"></a>\n<a href=\"https://github.com/earth-chris\"></a>\n\n[home-maxent]: https://biodiversityinformatics.amnh.org/open_source/maxent/\n[r-maxnet]: https://github.com/mrmaxent/maxnet\n[^1]: [EO Lab, Planet Labs PBC](https://www.planet.com)\n[^2]: [Center for Conservation Biology, Stanford University](https://ccb.stanford.edu)\n\n\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Species distribution modeling support tools",
"version": "1.0.1",
"split_keywords": [
"biogeography",
"ecology",
"conservation",
"sdm",
"species distribution modeling",
"maxent"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d8b547abac6f479bb532e2b5f352906e90beee3f74816e0f5acedc60819c266b",
"md5": "42357394bf2459f792c4e01b13661130",
"sha256": "0e222e1dcb05e36c9819a7d81748957466a9cf90b6834c854e4a930d54a952df"
},
"downloads": -1,
"filename": "elapid-1.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "42357394bf2459f792c4e01b13661130",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7.0",
"size": 78409,
"upload_time": "2023-04-10T08:12:54",
"upload_time_iso_8601": "2023-04-10T08:12:54.557975Z",
"url": "https://files.pythonhosted.org/packages/d8/b5/47abac6f479bb532e2b5f352906e90beee3f74816e0f5acedc60819c266b/elapid-1.0.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fe8d2e1ee247b1c8016a58f4bd3bcad910f240a7b03837fa04c6e2348acfcd80",
"md5": "3a675ea4e92d81851d54472680e30bf2",
"sha256": "9b34d8afccc1608140da19ea844bd22c25e52b53572a0af5e256323a5307d0dd"
},
"downloads": -1,
"filename": "elapid-1.0.1.tar.gz",
"has_sig": false,
"md5_digest": "3a675ea4e92d81851d54472680e30bf2",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7.0",
"size": 60149,
"upload_time": "2023-04-10T08:12:56",
"upload_time_iso_8601": "2023-04-10T08:12:56.210255Z",
"url": "https://files.pythonhosted.org/packages/fe/8d/2e1ee247b1c8016a58f4bd3bcad910f240a7b03837fa04c6e2348acfcd80/elapid-1.0.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-04-10 08:12:56",
"github": false,
"gitlab": false,
"bitbucket": false,
"lcname": "elapid"
}
JSON