| Name | ezmodel JSON |
| Version |
0.2.1
JSON |
| download |
| home_page | |
| Summary | Machine Learning, Model, Surrogate, Metamodels, Response Surface |
| upload_time | 2023-11-25 22:13:17 |
| maintainer | |
| docs_url | None |
| author | Julian Blank |
| requires_python | >=3.6 |
| license | Apache License 2.0 |
| keywords |
model
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
ezmodel - A Common Interface for Models and Model Selection
====================================================================
For more information about our toolbox, users are encouraged to read our documentation.
https://anyoptimization.com/projects/ezmodel/
|python| |license|
.. |python| image:: https://img.shields.io/badge/python-3.9-blue.svg
:alt: python 3.6
.. |license| image:: https://img.shields.io/badge/license-apache-orange.svg
:alt: license apache
:target: https://www.apache.org/licenses/LICENSE-2.0
Installation
====================================================================
The official release is always available at PyPi:
.. code:: bash
pip install -U ezmodel
Usage
==================================
Benchmarking
==================================
.. code:: python
import numpy as np
import pandas as pd
pd.set_option('display.expand_frame_repr', False)
pd.set_option('max_colwidth', 1000)
from ezmodel.core.benchmark import Benchmark
from ezmodel.core.factory import models_from_clazzes
from ezmodel.models.kriging import Kriging
from ezmodel.models.rbf import RBF
from ezmodel.util.partitioning.crossvalidation import CrossvalidationPartitioning
X = np.random.random((100, 3)) * 2 * np.pi
y = np.sin(X).sum(axis=1)
models = models_from_clazzes(RBF, Kriging)
# set up the benchmark and add the models to be used
benchmark = Benchmark(models, n_threads=4, verbose=True, raise_exception=True)
# create partitions to validate the performance of each model
partitions = CrossvalidationPartitioning(k_folds=5, seed=1).do(X)
# runs the experiment with the specified partitioning
benchmark.do(X, y, partitions=partitions)
# print out the benchmark results
print(benchmark.statistics("mae"))
::
mae
mean std min max median
label
Kriging[regr=constant,corr=gauss,thetaU=100,ARD=False] 0.017159 0.007472 0.009658 0.025359 0.014855
Kriging[regr=constant,corr=gauss,thetaU=20,ARD=False] 0.017159 0.007472 0.009658 0.025359 0.014855
Kriging[regr=linear,corr=gauss,thetaU=100,ARD=False] 0.018064 0.008069 0.010350 0.027456 0.014246
Kriging[regr=linear,corr=gauss,thetaU=20,ARD=False] 0.018064 0.008069 0.010350 0.027456 0.014246
Kriging[regr=constant,corr=gauss,thetaU=100,ARD=True] 0.021755 0.007409 0.011955 0.028896 0.025163
Kriging[regr=constant,corr=gauss,thetaU=20,ARD=True] 0.021755 0.007409 0.011955 0.028896 0.025163
Kriging[regr=linear,corr=gauss,thetaU=20,ARD=True] 0.025018 0.011348 0.011576 0.040585 0.022124
Kriging[regr=linear,corr=gauss,thetaU=100,ARD=True] 0.025018 0.011348 0.011576 0.040585 0.022124
Kriging[regr=constant,corr=exp,thetaU=100,ARD=False] 0.034493 0.009328 0.025092 0.045610 0.030661
Kriging[regr=constant,corr=exp,thetaU=20,ARD=False] 0.034493 0.009328 0.025092 0.045610 0.030661
Kriging[regr=linear,corr=exp,thetaU=100,ARD=False] 0.035734 0.009922 0.025611 0.047926 0.031473
Kriging[regr=linear,corr=exp,thetaU=20,ARD=False] 0.035734 0.009922 0.025611 0.047926 0.031473
Kriging[regr=constant,corr=exp,thetaU=100,ARD=True] 0.051527 0.010941 0.037944 0.065866 0.047440
Kriging[regr=constant,corr=exp,thetaU=20,ARD=True] 0.051527 0.010941 0.037944 0.065866 0.047440
Kriging[regr=linear,corr=exp,thetaU=100,ARD=True] 0.065867 0.025312 0.039058 0.104449 0.059957
Kriging[regr=linear,corr=exp,thetaU=20,ARD=True] 0.065867 0.025312 0.039058 0.104449 0.059957
RBF[kernel=cubic,tail=quadratic,normalized=True] 0.121947 0.033552 0.077895 0.167120 0.127345
RBF[kernel=cubic,tail=constant,normalized=True] 0.125348 0.037982 0.072579 0.169413 0.140753
RBF[kernel=cubic,tail=linear,normalized=True] 0.125474 0.038609 0.071268 0.169843 0.137987
RBF[kernel=cubic,tail=linear+quadratic,normalized=True] 0.126070 0.039773 0.071279 0.171862 0.135489
RBF
----------------------------------
.. code:: python
import matplotlib.pyplot as plt
import numpy as np
from ezmodel.models.rbf import RBF
from ezmodel.util.sample_from_func import sine_function
rbf = RBF(kernel="gaussian")
# create some data to test this model on
X, y, _X, _y = sine_function(20, 200)
# let the model fit the data
rbf.fit(X, y)
# predict the data using the model
y_hat = rbf.predict(_X)
# predict the data using the model
_X = _X[np.argsort(_X[:, 0])]
y_hat = rbf.predict(_X)
plt.scatter(X, y, label="Data")
plt.plot(_X, y_hat, color="black", label="RBF")
plt.legend()
plt.show()
Kriging
----------------------------------
.. code:: python
import matplotlib.pyplot as plt
import numpy as np
from ezmodel.models.kriging import Kriging
from ezmodel.util.sample_from_func import square_function
model = Kriging(regr="linear",
corr="gauss",
ARD=False)
# create some data to test this model on
X, y, _X, _y = square_function(100, 20)
# let the model fit the data
model.fit(X, y)
# predict the data using the model
y_hat = model.predict(_X)
# predict the data using the model
_X = _X[np.argsort(_X[:, 0])]
y_hat = model.predict(_X)
plt.scatter(X, y, label="Data")
plt.plot(_X, y_hat, color="black", label="RBF")
plt.legend()
plt.show()
Contact
=======
Feel free to contact us if you have any question:
::
Julian Blank (blankjul [at] msu.edu)
Michigan State University
Computational Optimization and Innovation Laboratory (COIN)
East Lansing, MI 48824, USA
Raw data
{
"_id": null,
"home_page": "",
"name": "ezmodel",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.6",
"maintainer_email": "",
"keywords": "model",
"author": "Julian Blank",
"author_email": "blankjul@msu.edu",
"download_url": "https://files.pythonhosted.org/packages/4c/d1/538d104e4377e1a3583ae9fc6a30e26f2b23b63755a9b14f2893d2be2278/ezmodel-0.2.1.tar.gz",
"platform": "any",
"description": "ezmodel - A Common Interface for Models and Model Selection\n====================================================================\n\nFor more information about our toolbox, users are encouraged to read our documentation.\nhttps://anyoptimization.com/projects/ezmodel/\n\n\n|python| |license|\n\n\n.. |python| image:: https://img.shields.io/badge/python-3.9-blue.svg\n :alt: python 3.6\n\n.. |license| image:: https://img.shields.io/badge/license-apache-orange.svg\n :alt: license apache\n :target: https://www.apache.org/licenses/LICENSE-2.0\n\n\n\nInstallation\n====================================================================\n\nThe official release is always available at PyPi:\n\n.. code:: bash\n\n pip install -U ezmodel\n\n\n\nUsage\n==================================\n\n\n\nBenchmarking\n==================================\n\n\n.. code:: python\n\n \n import numpy as np\n\n import pandas as pd\n pd.set_option('display.expand_frame_repr', False)\n pd.set_option('max_colwidth', 1000)\n\n from ezmodel.core.benchmark import Benchmark\n from ezmodel.core.factory import models_from_clazzes\n from ezmodel.models.kriging import Kriging\n from ezmodel.models.rbf import RBF\n from ezmodel.util.partitioning.crossvalidation import CrossvalidationPartitioning\n\n X = np.random.random((100, 3)) * 2 * np.pi\n y = np.sin(X).sum(axis=1)\n\n models = models_from_clazzes(RBF, Kriging)\n\n # set up the benchmark and add the models to be used\n benchmark = Benchmark(models, n_threads=4, verbose=True, raise_exception=True)\n\n # create partitions to validate the performance of each model\n partitions = CrossvalidationPartitioning(k_folds=5, seed=1).do(X)\n\n # runs the experiment with the specified partitioning\n benchmark.do(X, y, partitions=partitions)\n\n # print out the benchmark results\n print(benchmark.statistics(\"mae\"))\n\n\n\n::\n\n mae\n mean std min max median\n label\n Kriging[regr=constant,corr=gauss,thetaU=100,ARD=False] 0.017159 0.007472 0.009658 0.025359 0.014855\n Kriging[regr=constant,corr=gauss,thetaU=20,ARD=False] 0.017159 0.007472 0.009658 0.025359 0.014855\n Kriging[regr=linear,corr=gauss,thetaU=100,ARD=False] 0.018064 0.008069 0.010350 0.027456 0.014246\n Kriging[regr=linear,corr=gauss,thetaU=20,ARD=False] 0.018064 0.008069 0.010350 0.027456 0.014246\n Kriging[regr=constant,corr=gauss,thetaU=100,ARD=True] 0.021755 0.007409 0.011955 0.028896 0.025163\n Kriging[regr=constant,corr=gauss,thetaU=20,ARD=True] 0.021755 0.007409 0.011955 0.028896 0.025163\n Kriging[regr=linear,corr=gauss,thetaU=20,ARD=True] 0.025018 0.011348 0.011576 0.040585 0.022124\n Kriging[regr=linear,corr=gauss,thetaU=100,ARD=True] 0.025018 0.011348 0.011576 0.040585 0.022124\n Kriging[regr=constant,corr=exp,thetaU=100,ARD=False] 0.034493 0.009328 0.025092 0.045610 0.030661\n Kriging[regr=constant,corr=exp,thetaU=20,ARD=False] 0.034493 0.009328 0.025092 0.045610 0.030661\n Kriging[regr=linear,corr=exp,thetaU=100,ARD=False] 0.035734 0.009922 0.025611 0.047926 0.031473\n Kriging[regr=linear,corr=exp,thetaU=20,ARD=False] 0.035734 0.009922 0.025611 0.047926 0.031473\n Kriging[regr=constant,corr=exp,thetaU=100,ARD=True] 0.051527 0.010941 0.037944 0.065866 0.047440\n Kriging[regr=constant,corr=exp,thetaU=20,ARD=True] 0.051527 0.010941 0.037944 0.065866 0.047440\n Kriging[regr=linear,corr=exp,thetaU=100,ARD=True] 0.065867 0.025312 0.039058 0.104449 0.059957\n Kriging[regr=linear,corr=exp,thetaU=20,ARD=True] 0.065867 0.025312 0.039058 0.104449 0.059957\n RBF[kernel=cubic,tail=quadratic,normalized=True] 0.121947 0.033552 0.077895 0.167120 0.127345\n RBF[kernel=cubic,tail=constant,normalized=True] 0.125348 0.037982 0.072579 0.169413 0.140753\n RBF[kernel=cubic,tail=linear,normalized=True] 0.125474 0.038609 0.071268 0.169843 0.137987\n RBF[kernel=cubic,tail=linear+quadratic,normalized=True] 0.126070 0.039773 0.071279 0.171862 0.135489\n\n\n\n\nRBF\n----------------------------------\n\n\n.. code:: python\n\n \n import matplotlib.pyplot as plt\n import numpy as np\n\n from ezmodel.models.rbf import RBF\n from ezmodel.util.sample_from_func import sine_function\n\n rbf = RBF(kernel=\"gaussian\")\n\n # create some data to test this model on\n X, y, _X, _y = sine_function(20, 200)\n\n # let the model fit the data\n rbf.fit(X, y)\n\n # predict the data using the model\n y_hat = rbf.predict(_X)\n\n # predict the data using the model\n _X = _X[np.argsort(_X[:, 0])]\n y_hat = rbf.predict(_X)\n\n plt.scatter(X, y, label=\"Data\")\n plt.plot(_X, y_hat, color=\"black\", label=\"RBF\")\n plt.legend()\n plt.show()\n\n\n\nKriging\n----------------------------------\n\n\n.. code:: python\n\n \n import matplotlib.pyplot as plt\n import numpy as np\n\n from ezmodel.models.kriging import Kriging\n from ezmodel.util.sample_from_func import square_function\n\n model = Kriging(regr=\"linear\",\n corr=\"gauss\",\n ARD=False)\n\n # create some data to test this model on\n X, y, _X, _y = square_function(100, 20)\n\n # let the model fit the data\n model.fit(X, y)\n\n # predict the data using the model\n y_hat = model.predict(_X)\n\n # predict the data using the model\n _X = _X[np.argsort(_X[:, 0])]\n y_hat = model.predict(_X)\n\n plt.scatter(X, y, label=\"Data\")\n plt.plot(_X, y_hat, color=\"black\", label=\"RBF\")\n plt.legend()\n plt.show()\n\n\n\n\nContact\n=======\n\n\nFeel free to contact us if you have any question:\n\n::\n\n Julian Blank (blankjul [at] msu.edu)\n Michigan State University\n Computational Optimization and Innovation Laboratory (COIN)\n East Lansing, MI 48824, USA\n",
"bugtrack_url": null,
"license": "Apache License 2.0",
"summary": "Machine Learning, Model, Surrogate, Metamodels, Response Surface",
"version": "0.2.1",
"project_urls": null,
"split_keywords": [
"model"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "4cd1538d104e4377e1a3583ae9fc6a30e26f2b23b63755a9b14f2893d2be2278",
"md5": "83965d633ea60daa95e1ca61cd81d0f4",
"sha256": "e72d44b87cc13b7400240a9b56ad6ee4b447b0498b6a02829f77cf7afeee5a0d"
},
"downloads": -1,
"filename": "ezmodel-0.2.1.tar.gz",
"has_sig": false,
"md5_digest": "83965d633ea60daa95e1ca61cd81d0f4",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6",
"size": 27802,
"upload_time": "2023-11-25T22:13:17",
"upload_time_iso_8601": "2023-11-25T22:13:17.952489Z",
"url": "https://files.pythonhosted.org/packages/4c/d1/538d104e4377e1a3583ae9fc6a30e26f2b23b63755a9b14f2893d2be2278/ezmodel-0.2.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-11-25 22:13:17",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "ezmodel"
}
