| Name | DumME JSON |
| Version |
0.1.0
JSON |
| download |
| home_page | |
| Summary | Mixed Effects Dummy Model |
| upload_time | 2024-01-30 08:08:33 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | >=3.9 |
| license | |
| keywords |
random forest
machine learning
mixed effects
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
[](https://doi.org/10.5281/zenodo.10588922)
# DumME: Mixed Effects Dummy Model
This is an adaptation of MERF (https://github.com/manifoldai/merf). The main
difference is that this version is fully compliant with the scikit-learn API.
Other difference include:
- The name: MERF was renamed to the more general MixedEffectsModel
- The default fixed-effects model: dummy model instead of random forest
- The package structure: stripped down to its core and then upgraded to use
modern standards
- Test suite: using pytest instead of unittest
> [!CAUTION]
> We are currently not maintaining or developing this further. Ideally we would
> contribute our changes to the original version of MERF
> (see https://github.com/manifoldai/merf/issues/68).
> Do reach out if you want to build upon or collaborate with us on this.
## Using this version
Install via github:
```bash
pip install git+https://github.com/phenology/merf
```
Instantiate the dummy model:
```python
from dumme.dumme import MixedEffectsModel
from dumme.utils import DummeDataGenerator
# Get some sample data
dg = DummeDataGenerator(m=0.6, sigma_b=4.5, sigma_e=1)
df, _ = dg.generate_split_samples([1, 3], [3, 2], [1, 1])
y = df.pop("y")
x = df
# Fit a dummy model
# Notice the signature of the `fit` method: first X and y, and the other args are optional.
me_dummy = MixedEffectsModel()
me_dummy.fit(X, y)
# or
me_dummy.fit(X, y, cluster_column="cluster", fixed_effects=["X_0", "X_1", "X_2"], random_effects=["Z"])
# Predict only accepts X as input. It is assumed new data is structured
# in the same way as the original training data.
new_X = X.copy()
me_dummy.predict(new_X)
```
To get the "original" MERF (but still with the new fit signature):
```python
from sklearn.ensemble import RandomForestRegressor
rf = RandomForestRegressor(n_estimators=300, n_jobs=-1)
me_rf = MixedEffectsModel(rf)
me_rf.fit(X, y)
```
Raw data
{
"_id": null,
"home_page": "",
"name": "DumME",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": "",
"keywords": "random forest,machine learning,mixed effects",
"author": "",
"author_email": "\"Peter Kalverla.\" <p.kalverla@esciencecenter.nl>, \"Stefan Verhoeven.\" <s.verhoeven@esciencecenter.nl>",
"download_url": "https://files.pythonhosted.org/packages/df/d0/cca14c29ae9e3342453307a76d1551e0faab1abdecb8d073a4e85ff06f0e/DumME-0.1.0.tar.gz",
"platform": null,
"description": "[](https://doi.org/10.5281/zenodo.10588922)\n\n# DumME: Mixed Effects Dummy Model\n\nThis is an adaptation of MERF (https://github.com/manifoldai/merf). The main\ndifference is that this version is fully compliant with the scikit-learn API.\n\nOther difference include:\n\n- The name: MERF was renamed to the more general MixedEffectsModel\n- The default fixed-effects model: dummy model instead of random forest\n- The package structure: stripped down to its core and then upgraded to use\n modern standards\n- Test suite: using pytest instead of unittest\n\n> [!CAUTION]\n> We are currently not maintaining or developing this further. Ideally we would\n> contribute our changes to the original version of MERF\n> (see https://github.com/manifoldai/merf/issues/68).\n> Do reach out if you want to build upon or collaborate with us on this.\n\n## Using this version\n\nInstall via github:\n\n```bash\npip install git+https://github.com/phenology/merf\n```\n\nInstantiate the dummy model:\n\n```python\nfrom dumme.dumme import MixedEffectsModel\nfrom dumme.utils import DummeDataGenerator\n\n# Get some sample data\ndg = DummeDataGenerator(m=0.6, sigma_b=4.5, sigma_e=1)\ndf, _ = dg.generate_split_samples([1, 3], [3, 2], [1, 1])\ny = df.pop(\"y\")\nx = df\n\n# Fit a dummy model\n# Notice the signature of the `fit` method: first X and y, and the other args are optional.\nme_dummy = MixedEffectsModel()\nme_dummy.fit(X, y)\n\n# or\nme_dummy.fit(X, y, cluster_column=\"cluster\", fixed_effects=[\"X_0\", \"X_1\", \"X_2\"], random_effects=[\"Z\"])\n\n# Predict only accepts X as input. It is assumed new data is structured\n# in the same way as the original training data.\nnew_X = X.copy()\nme_dummy.predict(new_X)\n```\n\nTo get the \"original\" MERF (but still with the new fit signature):\n\n```python\nfrom sklearn.ensemble import RandomForestRegressor\n\nrf = RandomForestRegressor(n_estimators=300, n_jobs=-1)\nme_rf = MixedEffectsModel(rf)\nme_rf.fit(X, y)\n```\n",
"bugtrack_url": null,
"license": "",
"summary": "Mixed Effects Dummy Model",
"version": "0.1.0",
"project_urls": {
"Original": "https://github.com/manifoldai/merf",
"Repository": "https://github.com/phenology/dumme"
},
"split_keywords": [
"random forest",
"machine learning",
"mixed effects"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "9c02c54fc6a17222cc484bb70ff1b83793a5b54530891c7e6efd7a82340237b4",
"md5": "188ae823ca17be894e1c7226904e6644",
"sha256": "92684bee7ba9a8a0f1c35e49cdb90b55c6960874f769ae73a4a7a39b8f071c4a"
},
"downloads": -1,
"filename": "DumME-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "188ae823ca17be894e1c7226904e6644",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 12189,
"upload_time": "2024-01-30T08:08:31",
"upload_time_iso_8601": "2024-01-30T08:08:31.133236Z",
"url": "https://files.pythonhosted.org/packages/9c/02/c54fc6a17222cc484bb70ff1b83793a5b54530891c7e6efd7a82340237b4/DumME-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "dfd0cca14c29ae9e3342453307a76d1551e0faab1abdecb8d073a4e85ff06f0e",
"md5": "5886aa6a285b8333c47e184467a5f15c",
"sha256": "30b83ccfad3bc4cd2c2ed824226ef1efd1da205a03e102a7184d095a37941eab"
},
"downloads": -1,
"filename": "DumME-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "5886aa6a285b8333c47e184467a5f15c",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 14235,
"upload_time": "2024-01-30T08:08:33",
"upload_time_iso_8601": "2024-01-30T08:08:33.178284Z",
"url": "https://files.pythonhosted.org/packages/df/d0/cca14c29ae9e3342453307a76d1551e0faab1abdecb8d073a4e85ff06f0e/DumME-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-01-30 08:08:33",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "manifoldai",
"github_project": "merf",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"tox": true,
"lcname": "dumme"
}
