# auto-sktime
Automatic creation of time series forecasts, regression and classification.
## Installation
For trouble shooting and detailed installation instructions, see the documentation.
```
Operating system: Linux
Python version: Python 3.8, 3.9, and 3.10 (only 64 bit)
Package managers: pip
```
### pip
auto-sktime is available in pip. You can see all available wheels [here](https://test.pypi.org/project/auto-sktime).
```bash
pip install auto-sktime
```
or, with maximum dependencies,
```bash
pip install auto-sktime[all_extras]
```
## Remaining Useful Life Predictions (AutoRUL)
This section describes how to reproduce the results in the _AutoRUL_ paper. First, checkout the exact code that was used
to create the results. Therefore, you can use the tag [v0.1.0](https://github.com/Ennosigaeon/auto-sktime/tree/v0.1.0)
```bash
git checkout tags/v0.1.0 -b autorul
```
Next, switch to the `scripts` directory and use
```bash
python remaining_useful_lifetime.py <BENCHMARK>
```
to run a single benchmark data set. To view the available benchmarks and all configuration parameters run
```bash
python remaining_useful_lifetime.py --help
```
### Reproducing results
You can use the following commands to recreate the reported baseline results in the experiments of the paper.
```bash
python remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_lstm
python remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_cnn
python remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_transformer
python remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 7200 --multi_fidelity False --include baseline_rf
python remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 200 --timeout 7200 --multi_fidelity False --ensemble_size 1 --include baseline_svm
```
with `<BENCHMARK>` being one of `{cmapss,cmapss_1,cmapss_2,cmapss_3,cmapss_4,femto_bearing,filtration,phm08,phme20}`.
For the _AutoRUL_ evaluation only the benchmark is provided and all remaining default configurations are used.
```bash
python remaining_useful_lifetime.py <BENCHMARK>
```
To reproduce the results from AutoCoevoRUL, checkout the [repository](https://github.com/Ennosigaeon/AutoCoevoRUL) from
Github and use the [autocoevorul.py](scripts/autocoevorul.py) file to either export the data sets or import the results.
## Note
This project has been set up using PyScaffold 4.2.1. For details and usage
information on PyScaffold see https://pyscaffold.org/.
## Building
To create a new release of `auto-sktime` you will have to install `build` and `twine`
```bash
pip install build twine
python -m build
```
Raw data
{
"_id": null,
"home_page": "https://github.com/Ennosigaeon/auto-sktime/",
"name": "auto-sktime",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "",
"keywords": "",
"author": "Marc Zoeller",
"author_email": "marc.zoeller@usu.com",
"download_url": "https://files.pythonhosted.org/packages/79/4f/c2c2029ac54daee501a601a9d62bd2cc80eb283e43ac936f119d8710e6ff/auto-sktime-0.1.0.tar.gz",
"platform": "any",
"description": "# auto-sktime\n\nAutomatic creation of time series forecasts, regression and classification.\n\n## Installation\n\nFor trouble shooting and detailed installation instructions, see the documentation.\n\n```\nOperating system: Linux\nPython version: Python 3.8, 3.9, and 3.10 (only 64 bit)\nPackage managers: pip\n```\n\n### pip\n\nauto-sktime is available in pip. You can see all available wheels [here](https://test.pypi.org/project/auto-sktime).\n\n```bash\npip install auto-sktime\n```\n\nor, with maximum dependencies,\n\n```bash\npip install auto-sktime[all_extras]\n```\n\n## Remaining Useful Life Predictions (AutoRUL)\n\nThis section describes how to reproduce the results in the _AutoRUL_ paper. First, checkout the exact code that was used\nto create the results. Therefore, you can use the tag [v0.1.0](https://github.com/Ennosigaeon/auto-sktime/tree/v0.1.0)\n\n```bash\ngit checkout tags/v0.1.0 -b autorul\n```\n\nNext, switch to the `scripts` directory and use\n\n```bash\npython remaining_useful_lifetime.py <BENCHMARK>\n```\n\nto run a single benchmark data set. To view the available benchmarks and all configuration parameters run\n\n```bash\npython remaining_useful_lifetime.py --help\n```\n\n### Reproducing results\n\nYou can use the following commands to recreate the reported baseline results in the experiments of the paper.\n\n```bash\npython remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_lstm\npython remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_cnn\npython remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 3600 --multi_fidelity False --include baseline_transformer\npython remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 1 --timeout 7200 --multi_fidelity False --include baseline_rf\npython remaining_useful_lifetime.py <BENCHMARK> --runcount_limit 200 --timeout 7200 --multi_fidelity False --ensemble_size 1 --include baseline_svm\n```\n\nwith `<BENCHMARK>` being one of `{cmapss,cmapss_1,cmapss_2,cmapss_3,cmapss_4,femto_bearing,filtration,phm08,phme20}`.\nFor the _AutoRUL_ evaluation only the benchmark is provided and all remaining default configurations are used.\n\n```bash\npython remaining_useful_lifetime.py <BENCHMARK>\n```\n\nTo reproduce the results from AutoCoevoRUL, checkout the [repository](https://github.com/Ennosigaeon/AutoCoevoRUL) from\nGithub and use the [autocoevorul.py](scripts/autocoevorul.py) file to either export the data sets or import the results.\n\n## Note\n\nThis project has been set up using PyScaffold 4.2.1. For details and usage\ninformation on PyScaffold see https://pyscaffold.org/.\n\n## Building\n\nTo create a new release of `auto-sktime` you will have to install `build` and `twine`\n\n```bash\npip install build twine\npython -m build\n\n```\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Automatic creation of time series forecasts, regression and classification",
"version": "0.1.0",
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "2c034f8829d16c6c9d31f5ad8b0d4c6d805aea9088a72fbdc74ea506e17e0e33",
"md5": "7c976041f1986493b7171064e18fbb0c",
"sha256": "a52e8ad2837bc419b8f3308d3849d2887bede31fc20bae38098dc162dae1d422"
},
"downloads": -1,
"filename": "auto_sktime-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "7c976041f1986493b7171064e18fbb0c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 21337669,
"upload_time": "2023-03-28T22:02:10",
"upload_time_iso_8601": "2023-03-28T22:02:10.124888Z",
"url": "https://files.pythonhosted.org/packages/2c/03/4f8829d16c6c9d31f5ad8b0d4c6d805aea9088a72fbdc74ea506e17e0e33/auto_sktime-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "794fc2c2029ac54daee501a601a9d62bd2cc80eb283e43ac936f119d8710e6ff",
"md5": "c99de22a7a4ef9fe7959ec3780bc4a38",
"sha256": "1d0976f64db26298fb71ff2324e6cd4f091910c38648d2fcf1c13c3bc9b0af32"
},
"downloads": -1,
"filename": "auto-sktime-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "c99de22a7a4ef9fe7959ec3780bc4a38",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 20664983,
"upload_time": "2023-03-28T22:02:20",
"upload_time_iso_8601": "2023-03-28T22:02:20.822996Z",
"url": "https://files.pythonhosted.org/packages/79/4f/c2c2029ac54daee501a601a9d62bd2cc80eb283e43ac936f119d8710e6ff/auto-sktime-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-03-28 22:02:20",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "Ennosigaeon",
"github_project": "auto-sktime",
"travis_ci": false,
"coveralls": true,
"github_actions": false,
"tox": true,
"lcname": "auto-sktime"
}
JSON
