| Name | discontinuum JSON |
| Version |
1.0.7
JSON |
| download |
| home_page | None |
| Summary | Estimate discontinuous timeseries from continuous covariates. |
| upload_time | 2025-07-22 18:45:16 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.9 |
| license | License
=======
Unless otherwise noted, this project is in the public domain in the United
States because it contains materials that originally came from the United
States Geological Survey, an agency of the United States Department of
Interior. For more information, see the official USGS copyright policy at
https://www.usgs.gov/information-policies-and-instructions/copyrights-and-credits
Additionally, we waive copyright and related rights in the work
worldwide through the CC0 1.0 Universal public domain dedication.
CC0 1.0 Universal Summary
-------------------------
This is a human-readable summary of the
[Legal Code (read the full text)][1].
### No Copyright
The person who associated a work with this deed has dedicated the work to
the public domain by waiving all of his or her rights to the work worldwide
under copyright law, including all related and neighboring rights, to the
extent allowed by law.
You can copy, modify, distribute and perform the work, even for commercial
purposes, all without asking permission.
### Other Information
In no way are the patent or trademark rights of any person affected by CC0,
nor are the rights that other persons may have in the work or in how the
work is used, such as publicity or privacy rights.
Unless expressly stated otherwise, the person who associated a work with
this deed makes no warranties about the work, and disclaims liability for
all uses of the work, to the fullest extent permitted by applicable law.
When using or citing the work, you should not imply endorsement by the
author or the affirmer.
[1]: https://creativecommons.org/publicdomain/zero/1.0/legalcode
|
| keywords |
timeseries
gaussian processes
discontinuous
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# discontinuum
> [!WARNING]
> Experimental.
## Overview
`discontinuum` is a middleware for developing Gaussian process (GP) models.
GP's are a flexible approach to machine learning,
which are naturally suited for applications with sparse and noisy data
or for uncertainty analysis.
However, fitting GP's is numerically expensive,
which has led to a range of optimizations with different tradeoffs.
Ideally, we could quickly write mathematical models, then run them on whichever
"engine" is best suited for a particular problem.
Most model applications also include a fair amount of "boiler plate"
in the form of utility functions for plotting, managing metadata, data pre-processing, etc.
`discontinum` packages several engines and helper utilities into a single ecosystem
to simplify the processes of prototyping GP models.
## Installation
```
pip install discontinuum
```
## Models
### loadset-gp
`loadest-gp` is Gaussian-process model for estimating river constituent time series,
which borrows its namesake from the venerable LOAD ESTimator (LOADEST) software program.
However, LOADEST has several serious limitations
---it's essentially a linear regression---and it has been all but replaced by
the more flexible Weighted Regression on Time Discharge and Season (WRTDS),
which allows the relation between target and covariate to vary through time.
`loadest-gp` takes the WRTDS idea and reimplements it as a GP.
github/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/loadest-gp-demo.ipynb
Try it out in the [loadest-gp demo](https://github.com/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/loadest-gp-demo.ipynb).
### rating-gp
`rating-gp` is a Gaussian-process model for estimating river flow from stage time series.
Try it out in the [rating-gp demo](https://github.com/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/rating-gp-demo.ipynb).
## Engines
Currently, the only supported engines are the marginal likelihood implementation in `pymc` and `gpytorch`.
Latent GP implementations could be added in the future.
In general, the `gpytorch` implementation is faster and provides a lot a powerful features,
like GPU support, whereas `pymc` is a more complete probabilistic-programming framework,
which can be "friendlier" for certain use cases.
Porting a model from one engine to another usually involves completely rewriting the guts of the GP.
but any other boiler-plate functions should be unaffected.
## Roadmap
```mermaid
mindmap
root((discontinuum))
data providers
USGS
etc
engines
PyMC
PyTorch
utilities
pre-processing
post-processing
plotting
models
loadest-gp
rating-gp
```
Raw data
{
"_id": null,
"home_page": null,
"name": "discontinuum",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": "Timothy Hodson <thodson@usgs.gov>",
"keywords": "timeseries, Gaussian processes, discontinuous",
"author": null,
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/0e/77/2000863d44071bc255e9136b89fa4ff4525ccf84e70bc724f0c60ddf1fbb/discontinuum-1.0.7.tar.gz",
"platform": null,
"description": "# discontinuum\n\n\n\n> [!WARNING] \n> Experimental.\n\n## Overview\n`discontinuum` is a middleware for developing Gaussian process (GP) models.\n\nGP's are a flexible approach to machine learning,\nwhich are naturally suited for applications with sparse and noisy data\nor for uncertainty analysis.\nHowever, fitting GP's is numerically expensive,\nwhich has led to a range of optimizations with different tradeoffs.\nIdeally, we could quickly write mathematical models, then run them on whichever\n\"engine\" is best suited for a particular problem.\n\nMost model applications also include a fair amount of \"boiler plate\"\nin the form of utility functions for plotting, managing metadata, data pre-processing, etc.\n`discontinum` packages several engines and helper utilities into a single ecosystem\nto simplify the processes of prototyping GP models.\n\n## Installation\n```\npip install discontinuum\n```\n\n## Models\n\n### loadset-gp\n`loadest-gp` is Gaussian-process model for estimating river constituent time series,\nwhich borrows its namesake from the venerable LOAD ESTimator (LOADEST) software program.\nHowever, LOADEST has several serious limitations\n---it's essentially a linear regression---and it has been all but replaced by\nthe more flexible Weighted Regression on Time Discharge and Season (WRTDS),\nwhich allows the relation between target and covariate to vary through time.\n`loadest-gp` takes the WRTDS idea and reimplements it as a GP.\ngithub/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/loadest-gp-demo.ipynb\nTry it out in the [loadest-gp demo](https://github.com/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/loadest-gp-demo.ipynb).\n\n### rating-gp\n`rating-gp` is a Gaussian-process model for estimating river flow from stage time series.\nTry it out in the [rating-gp demo](https://github.com/thodson-usgs/discontinuum/blob/main/docs/source/notebooks/rating-gp-demo.ipynb).\n\n## Engines\nCurrently, the only supported engines are the marginal likelihood implementation in `pymc` and `gpytorch`.\nLatent GP implementations could be added in the future.\nIn general, the `gpytorch` implementation is faster and provides a lot a powerful features,\nlike GPU support, whereas `pymc` is a more complete probabilistic-programming framework,\nwhich can be \"friendlier\" for certain use cases.\n\nPorting a model from one engine to another usually involves completely rewriting the guts of the GP.\nbut any other boiler-plate functions should be unaffected.\n\n\n## Roadmap\n```mermaid\nmindmap\n root((discontinuum))\n data providers\n USGS\n etc\n engines\n PyMC\n PyTorch\n utilities\n pre-processing\n post-processing\n plotting\n models\n loadest-gp\n rating-gp\n```\n",
"bugtrack_url": null,
"license": "License\n =======\n \n Unless otherwise noted, this project is in the public domain in the United\n States because it contains materials that originally came from the United\n States Geological Survey, an agency of the United States Department of\n Interior. For more information, see the official USGS copyright policy at\n https://www.usgs.gov/information-policies-and-instructions/copyrights-and-credits\n \n Additionally, we waive copyright and related rights in the work\n worldwide through the CC0 1.0 Universal public domain dedication.\n \n \n CC0 1.0 Universal Summary\n -------------------------\n \n This is a human-readable summary of the\n [Legal Code (read the full text)][1].\n \n \n ### No Copyright\n \n The person who associated a work with this deed has dedicated the work to\n the public domain by waiving all of his or her rights to the work worldwide\n under copyright law, including all related and neighboring rights, to the\n extent allowed by law.\n \n You can copy, modify, distribute and perform the work, even for commercial\n purposes, all without asking permission.\n \n \n ### Other Information\n \n In no way are the patent or trademark rights of any person affected by CC0,\n nor are the rights that other persons may have in the work or in how the\n work is used, such as publicity or privacy rights.\n \n Unless expressly stated otherwise, the person who associated a work with\n this deed makes no warranties about the work, and disclaims liability for\n all uses of the work, to the fullest extent permitted by applicable law.\n When using or citing the work, you should not imply endorsement by the\n author or the affirmer.\n \n \n \n [1]: https://creativecommons.org/publicdomain/zero/1.0/legalcode\n ",
"summary": "Estimate discontinuous timeseries from continuous covariates.",
"version": "1.0.7",
"project_urls": {
"homepage": "https://github.com/thodson-usgs/discontinuum",
"repository": "https://github.com/thodson-usgs/discontinuum.git"
},
"split_keywords": [
"timeseries",
" gaussian processes",
" discontinuous"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "ebd406db79a8a72099e04b3a63f569cf778287b7a8499b8bbda2c1aa4b6b4462",
"md5": "bfabdf90cabc42316023bb9b7cddb364",
"sha256": "3500b0fb0966e36ea846760c68f6bf3af4064e31ae6781c7a10f51a26479ddd8"
},
"downloads": -1,
"filename": "discontinuum-1.0.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "bfabdf90cabc42316023bb9b7cddb364",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 41325,
"upload_time": "2025-07-22T18:45:15",
"upload_time_iso_8601": "2025-07-22T18:45:15.730381Z",
"url": "https://files.pythonhosted.org/packages/eb/d4/06db79a8a72099e04b3a63f569cf778287b7a8499b8bbda2c1aa4b6b4462/discontinuum-1.0.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "0e772000863d44071bc255e9136b89fa4ff4525ccf84e70bc724f0c60ddf1fbb",
"md5": "f0fc35bf9bfa1f0e142a2a0eb2f9acf6",
"sha256": "ec3232baeaa0da52775bb5dab1563c1ab30be7847e549f106b145828321c68cb"
},
"downloads": -1,
"filename": "discontinuum-1.0.7.tar.gz",
"has_sig": false,
"md5_digest": "f0fc35bf9bfa1f0e142a2a0eb2f9acf6",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 140837,
"upload_time": "2025-07-22T18:45:16",
"upload_time_iso_8601": "2025-07-22T18:45:16.658172Z",
"url": "https://files.pythonhosted.org/packages/0e/77/2000863d44071bc255e9136b89fa4ff4525ccf84e70bc724f0c60ddf1fbb/discontinuum-1.0.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-22 18:45:16",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "thodson-usgs",
"github_project": "discontinuum",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "discontinuum"
}
