{ "info": { "author": "Ghislain Vieilledent", "author_email": "ghislain.vieilledent@cirad.fr", "bugtrack_url": null, "classifiers": [ "Development Status :: 4 - Beta", "License :: OSI Approved :: GNU General Public License v3 (GPLv3)", "Operating System :: OS Independent", "Programming Language :: Python :: 2", "Programming Language :: Python :: 3", "Topic :: Scientific/Engineering :: Bio-Informatics" ], "description": "# forestatrisk Python package\n\n## Estimating the risk of deforestation in tropical countries\n\n`forestatrisk` is a Python package for estimating the spatial\nprobability of deforestation in the tropics depending on various\nspatial environmental variables.\n\nSpatial environmental variables can be derived from topography\n(altitude, slope, and aspect), accessibility (distance to roads,\ntowns, and forest edge), deforestation history (distance to previous\ndeforestation) or landscape management (location inside a protected area)\nfor example.\n\n### Sampling\n\nFunction `.sample()` allows the random sampling of observations points\nconsidering historical deforestation maps. The sampling is balanced\nand stratified considering remaining forest and deforested areas after\na given period of time. The function also retrieves information from\nenvironmental variables for each sampled point. The sampling is done\nby block to allow the computation on large study areas (e.g. country\nor continental scale) with a high spatial resolution (e.g. 30m).\n\n### Modelling\n\nFunction `.model_binomial_iCAR()` can be used to fit the deforestation\nmodel from the data. A linear Binomial logistic regression model is\nused to estimate the parameters of the deforestation model. The model\nincludes an intrinsic Conditional Autoregressive (iCAR) process to\naccount for the spatial autocorrelation of the observations\n(Vieilledent et al. 2014). Parameter inference is done in a\nhierarchical Bayesian framework. The function calls a Gibbs sampler\nwith a Metropolis algorithm written in pure C code to reduce\ncomputation time.\n\n### Predicting\n\nFunction `.predict()` allows predicting the deforestation probability\non the whole study area using the deforestation model fitted with the\n`.model()` function. The prediction is done by block to allow the\ncomputation on large study areas (e.g. country or continental scale)\nwith a high spatial resolution (e.g. 30m).\n\nFunction `.deforest()` predicts the future forest cover map based on a\nraster of probability of deforestation (rescaled from 1 to 65535),\nwhich is obtained from function `.predict()`, and an area (in\nhectares) to be deforested.\n\n## Tutorial\n\nWe wrote a tutorial using a Jupyter/IPython notebook to show how to\nuse the `forestatrisk` Python package. We took Madagascar as a case\nstudy considering past deforestation on the period 2000-2010,\nestimating deforestation probability for the year 2010, and projecting\nthe future forest cover in 2050. The notebook is available at the\nfollowing web adress: https://ecology.ghislainv.fr/forestatrisk\n\n## Reference\n\n**Vieilledent G., C. Merow, J. Gu\u00c3\u00a9lat, A. M. Latimer, M. K\u00c3\u00a9ry,\nA. E. Gelfand, A. M. Wilson, F. Mortier and J. A. Silander\nJr.** 2014. hSDM CRAN release v1.4 for hierarchical Bayesian species\ndistribution models. _Zenodo_.\ndoi: [10.5281/zenodo.48470](http://doi.org/10.5281/zenodo.48470)\n\n## Installation\n\nThe easiest way to install the `forestatrisk` Python package is via [pip](https://pip.pypa.io/en/stable/):\n\n```\n~$ sudo pip install --upgrade https://github.com/ghislainv/forestatrisk/archive/master.zip\n```\n\nbut you can also install it executing the `setup.py` file:\n\n```\n~$ git clone https://github.com/ghislainv/forestatrisk\n~$ cd forestatrisk\n~/forestatrisk$ sudo python setup.py install\n```\n\n## Figure\n\nMap of the probability of deforestation in Madagascar for the year\n2010 obtained with `forestatrisk`. Dark red: high probability of\ndeforestation, Dark green: low probability of deforestation.\n\n\n\n\n\n", "description_content_type": "text/markdown", "docs_url": null, "download_url": "", "downloads": { "last_day": -1, "last_month": -1, "last_week": -1 }, "home_page": "https://github.com/ghislainv/forestatrisk", "keywords": "deforestation hsdm hierarchical logistic model probability risk Bayesian spatial autocorrelation", "license": "GPLv3", "maintainer": "", "maintainer_email": "", "name": "forestatrisk", "package_url": "https://pypi.org/project/forestatrisk/", "platform": "", "project_url": "https://pypi.org/project/forestatrisk/", "project_urls": { "Homepage": "https://github.com/ghislainv/forestatrisk" }, "release_url": "https://pypi.org/project/forestatrisk/0.1/", "requires_dist": [ "numpy", "patsy", "scipy", "sklearn", "pandas", "matplotlib" ], "requires_python": ">=2.7", "summary": "This is the Python 'forestatrisk' package", "version": "0.1" }, "last_serial": 5568022, "releases": { "0.1": [ { "comment_text": "", "digests": { "md5": "d34d37781fc5bc25fe37666a13812858", "sha256": "19ca851aa5f5246705f4253d87835d4c986eada8766307e9fe9785ae79c8810c" }, "downloads": -1, "filename": "forestatrisk-0.1-cp37-cp37m-win_amd64.whl", "has_sig": false, "md5_digest": "d34d37781fc5bc25fe37666a13812858", "packagetype": "bdist_wheel", "python_version": "cp37", "requires_python": ">=2.7", "size": 111594, "upload_time": "2019-07-22T15:43:10", "url": "https://files.pythonhosted.org/packages/3b/9c/1fac80962ee5071c306b0c2607a18bfc7f58402a860c209e68b0215ae9f9/forestatrisk-0.1-cp37-cp37m-win_amd64.whl" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "d34d37781fc5bc25fe37666a13812858", "sha256": "19ca851aa5f5246705f4253d87835d4c986eada8766307e9fe9785ae79c8810c" }, "downloads": -1, "filename": "forestatrisk-0.1-cp37-cp37m-win_amd64.whl", "has_sig": false, "md5_digest": "d34d37781fc5bc25fe37666a13812858", "packagetype": "bdist_wheel", "python_version": "cp37", "requires_python": ">=2.7", "size": 111594, "upload_time": "2019-07-22T15:43:10", "url": "https://files.pythonhosted.org/packages/3b/9c/1fac80962ee5071c306b0c2607a18bfc7f58402a860c209e68b0215ae9f9/forestatrisk-0.1-cp37-cp37m-win_amd64.whl" } ] }