.. DO NOT EDIT. .. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY. .. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE: .. "examples/mlearning/regression_model/plot_random_forest_regression.py" .. LINE NUMBERS ARE GIVEN BELOW. .. only:: html .. note:: :class: sphx-glr-download-link-note :ref:`Go to the end ` to download the full example code. .. rst-class:: sphx-glr-example-title .. _sphx_glr_examples_mlearning_regression_model_plot_random_forest_regression.py: Random forest ============= A :class:`.RandomForestRegressor` is a random forest model based on `scikit-learn `__. .. GENERATED FROM PYTHON SOURCE LINES 28-39 .. code-block:: Python from __future__ import annotations from matplotlib import pyplot as plt from numpy import array from gemseo import create_design_space from gemseo import create_discipline from gemseo import sample_disciplines from gemseo.mlearning import create_regression_model .. GENERATED FROM PYTHON SOURCE LINES 40-45 Problem ------- In this example, we represent the function :math:`f(x)=(6x-2)^2\sin(12x-4)` :cite:`forrester2008` by the :class:`.AnalyticDiscipline` .. GENERATED FROM PYTHON SOURCE LINES 45-50 .. code-block:: Python discipline = create_discipline( "AnalyticDiscipline", name="f", expressions={"y": "(6*x-2)**2*sin(12*x-4)"}, ) .. GENERATED FROM PYTHON SOURCE LINES 51-52 and seek to approximate it over the input space .. GENERATED FROM PYTHON SOURCE LINES 52-55 .. code-block:: Python input_space = create_design_space() input_space.add_variable("x", lower_bound=0.0, upper_bound=1.0) .. GENERATED FROM PYTHON SOURCE LINES 56-58 To do this, we create a training dataset with 6 equispaced points: .. GENERATED FROM PYTHON SOURCE LINES 58-62 .. code-block:: Python training_dataset = sample_disciplines( [discipline], input_space, "y", algo_name="PYDOE_FULLFACT", n_samples=6 ) .. rst-class:: sphx-glr-script-out .. code-block:: none INFO - 16:19:15: *** Start Sampling execution *** INFO - 16:19:15: Sampling INFO - 16:19:15: Disciplines: f INFO - 16:19:15: MDO formulation: MDF INFO - 16:19:15: Running the algorithm PYDOE_FULLFACT: INFO - 16:19:15: 17%|█▋ | 1/6 [00:00<00:00, 554.80 it/sec] INFO - 16:19:15: 33%|███▎ | 2/6 [00:00<00:00, 842.48 it/sec] INFO - 16:19:15: 50%|█████ | 3/6 [00:00<00:00, 1053.67 it/sec] INFO - 16:19:15: 67%|██████▋ | 4/6 [00:00<00:00, 1193.94 it/sec] INFO - 16:19:15: 83%|████████▎ | 5/6 [00:00<00:00, 1307.37 it/sec] INFO - 16:19:15: 100%|██████████| 6/6 [00:00<00:00, 1362.60 it/sec] INFO - 16:19:15: *** End Sampling execution *** .. GENERATED FROM PYTHON SOURCE LINES 63-69 Basics ------ Training ~~~~~~~~ Then, we train an random forest regression model from these samples: .. GENERATED FROM PYTHON SOURCE LINES 69-72 .. code-block:: Python model = create_regression_model("RandomForestRegressor", training_dataset) model.learn() .. GENERATED FROM PYTHON SOURCE LINES 73-77 Prediction ~~~~~~~~~~ Once it is built, we can predict the output value of :math:`f` at a new input point: .. GENERATED FROM PYTHON SOURCE LINES 77-81 .. code-block:: Python input_value = {"x": array([0.65])} output_value = model.predict(input_value) output_value .. rst-class:: sphx-glr-script-out .. code-block:: none {'y': array([-0.88837697])} .. GENERATED FROM PYTHON SOURCE LINES 82-83 but cannot predict its Jacobian value: .. GENERATED FROM PYTHON SOURCE LINES 83-88 .. code-block:: Python try: model.predict_jacobian(input_value) except NotImplementedError: print("The derivatives are not available for RandomForestRegressor.") .. rst-class:: sphx-glr-script-out .. code-block:: none The derivatives are not available for RandomForestRegressor. .. GENERATED FROM PYTHON SOURCE LINES 89-93 Plotting ~~~~~~~~ You can see that the random forest model is pretty good on the left, but bad on the right: .. GENERATED FROM PYTHON SOURCE LINES 93-105 .. code-block:: Python test_dataset = sample_disciplines( [discipline], input_space, "y", algo_name="PYDOE_FULLFACT", n_samples=100 ) input_data = test_dataset.get_view(variable_names=model.input_names).to_numpy() reference_output_data = test_dataset.get_view(variable_names="y").to_numpy().ravel() predicted_output_data = model.predict(input_data).ravel() plt.plot(input_data.ravel(), reference_output_data, label="Reference") plt.plot(input_data.ravel(), predicted_output_data, label="Regression - Basics") plt.grid() plt.legend() plt.show() .. image-sg:: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_001.png :alt: plot random forest regression :srcset: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_001.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-script-out .. code-block:: none INFO - 16:19:15: *** Start Sampling execution *** INFO - 16:19:15: Sampling INFO - 16:19:15: Disciplines: f INFO - 16:19:15: MDO formulation: MDF INFO - 16:19:15: Running the algorithm PYDOE_FULLFACT: INFO - 16:19:15: 1%| | 1/100 [00:00<00:00, 2908.67 it/sec] INFO - 16:19:15: 2%|▏ | 2/100 [00:00<00:00, 2244.74 it/sec] INFO - 16:19:15: 3%|▎ | 3/100 [00:00<00:00, 2179.99 it/sec] INFO - 16:19:15: 4%|▍ | 4/100 [00:00<00:00, 2144.33 it/sec] INFO - 16:19:15: 5%|▌ | 5/100 [00:00<00:00, 2127.36 it/sec] INFO - 16:19:15: 6%|▌ | 6/100 [00:00<00:00, 2132.70 it/sec] INFO - 16:19:15: 7%|▋ | 7/100 [00:00<00:00, 2096.25 it/sec] INFO - 16:19:15: 8%|▊ | 8/100 [00:00<00:00, 2108.62 it/sec] INFO - 16:19:15: 9%|▉ | 9/100 [00:00<00:00, 2086.03 it/sec] INFO - 16:19:15: 10%|█ | 10/100 [00:00<00:00, 2097.89 it/sec] INFO - 16:19:15: 11%|█ | 11/100 [00:00<00:00, 2087.85 it/sec] INFO - 16:19:15: 12%|█▏ | 12/100 [00:00<00:00, 2097.68 it/sec] INFO - 16:19:15: 13%|█▎ | 13/100 [00:00<00:00, 2093.85 it/sec] INFO - 16:19:15: 14%|█▍ | 14/100 [00:00<00:00, 2102.86 it/sec] INFO - 16:19:15: 15%|█▌ | 15/100 [00:00<00:00, 2102.06 it/sec] INFO - 16:19:15: 16%|█▌ | 16/100 [00:00<00:00, 2100.96 it/sec] INFO - 16:19:15: 17%|█▋ | 17/100 [00:00<00:00, 2096.72 it/sec] INFO - 16:19:15: 18%|█▊ | 18/100 [00:00<00:00, 2104.17 it/sec] INFO - 16:19:15: 19%|█▉ | 19/100 [00:00<00:00, 2103.46 it/sec] INFO - 16:19:15: 20%|██ | 20/100 [00:00<00:00, 2103.52 it/sec] INFO - 16:19:15: 21%|██ | 21/100 [00:00<00:00, 2101.15 it/sec] INFO - 16:19:15: 22%|██▏ | 22/100 [00:00<00:00, 2104.76 it/sec] INFO - 16:19:15: 23%|██▎ | 23/100 [00:00<00:00, 2096.29 it/sec] INFO - 16:19:15: 24%|██▍ | 24/100 [00:00<00:00, 2101.57 it/sec] INFO - 16:19:15: 25%|██▌ | 25/100 [00:00<00:00, 2093.63 it/sec] INFO - 16:19:15: 26%|██▌ | 26/100 [00:00<00:00, 2098.65 it/sec] INFO - 16:19:15: 27%|██▋ | 27/100 [00:00<00:00, 2094.44 it/sec] INFO - 16:19:15: 28%|██▊ | 28/100 [00:00<00:00, 2099.59 it/sec] INFO - 16:19:15: 29%|██▉ | 29/100 [00:00<00:00, 2100.81 it/sec] INFO - 16:19:15: 30%|███ | 30/100 [00:00<00:00, 2100.48 it/sec] INFO - 16:19:15: 31%|███ | 31/100 [00:00<00:00, 2101.39 it/sec] INFO - 16:19:15: 32%|███▏ | 32/100 [00:00<00:00, 2083.80 it/sec] INFO - 16:19:15: 33%|███▎ | 33/100 [00:00<00:00, 2079.23 it/sec] INFO - 16:19:15: 34%|███▍ | 34/100 [00:00<00:00, 2080.69 it/sec] INFO - 16:19:15: 35%|███▌ | 35/100 [00:00<00:00, 2081.72 it/sec] INFO - 16:19:15: 36%|███▌ | 36/100 [00:00<00:00, 2081.86 it/sec] INFO - 16:19:15: 37%|███▋ | 37/100 [00:00<00:00, 2082.94 it/sec] INFO - 16:19:15: 38%|███▊ | 38/100 [00:00<00:00, 2083.34 it/sec] INFO - 16:19:15: 39%|███▉ | 39/100 [00:00<00:00, 2084.38 it/sec] INFO - 16:19:15: 40%|████ | 40/100 [00:00<00:00, 2085.91 it/sec] INFO - 16:19:15: 41%|████ | 41/100 [00:00<00:00, 2086.41 it/sec] INFO - 16:19:15: 42%|████▏ | 42/100 [00:00<00:00, 2085.36 it/sec] INFO - 16:19:15: 43%|████▎ | 43/100 [00:00<00:00, 2088.14 it/sec] INFO - 16:19:15: 44%|████▍ | 44/100 [00:00<00:00, 2085.85 it/sec] INFO - 16:19:15: 45%|████▌ | 45/100 [00:00<00:00, 2088.75 it/sec] INFO - 16:19:15: 46%|████▌ | 46/100 [00:00<00:00, 2088.14 it/sec] INFO - 16:19:15: 47%|████▋ | 47/100 [00:00<00:00, 2089.08 it/sec] INFO - 16:19:15: 48%|████▊ | 48/100 [00:00<00:00, 2088.08 it/sec] INFO - 16:19:15: 49%|████▉ | 49/100 [00:00<00:00, 2091.26 it/sec] INFO - 16:19:15: 50%|█████ | 50/100 [00:00<00:00, 2086.84 it/sec] INFO - 16:19:15: 51%|█████ | 51/100 [00:00<00:00, 2089.31 it/sec] INFO - 16:19:15: 52%|█████▏ | 52/100 [00:00<00:00, 2086.36 it/sec] INFO - 16:19:15: 53%|█████▎ | 53/100 [00:00<00:00, 2088.76 it/sec] INFO - 16:19:15: 54%|█████▍ | 54/100 [00:00<00:00, 2087.05 it/sec] INFO - 16:19:15: 55%|█████▌ | 55/100 [00:00<00:00, 2089.71 it/sec] INFO - 16:19:15: 56%|█████▌ | 56/100 [00:00<00:00, 2089.95 it/sec] INFO - 16:19:15: 57%|█████▋ | 57/100 [00:00<00:00, 2090.82 it/sec] INFO - 16:19:15: 58%|█████▊ | 58/100 [00:00<00:00, 2091.17 it/sec] INFO - 16:19:15: 59%|█████▉ | 59/100 [00:00<00:00, 2091.57 it/sec] INFO - 16:19:15: 60%|██████ | 60/100 [00:00<00:00, 2091.73 it/sec] INFO - 16:19:15: 61%|██████ | 61/100 [00:00<00:00, 2092.93 it/sec] INFO - 16:19:15: 62%|██████▏ | 62/100 [00:00<00:00, 2093.13 it/sec] INFO - 16:19:15: 63%|██████▎ | 63/100 [00:00<00:00, 2093.63 it/sec] INFO - 16:19:15: 64%|██████▍ | 64/100 [00:00<00:00, 2095.81 it/sec] INFO - 16:19:15: 65%|██████▌ | 65/100 [00:00<00:00, 2095.59 it/sec] INFO - 16:19:15: 66%|██████▌ | 66/100 [00:00<00:00, 2096.82 it/sec] INFO - 16:19:15: 67%|██████▋ | 67/100 [00:00<00:00, 2094.53 it/sec] INFO - 16:19:15: 68%|██████▊ | 68/100 [00:00<00:00, 2096.35 it/sec] INFO - 16:19:15: 69%|██████▉ | 69/100 [00:00<00:00, 2094.21 it/sec] INFO - 16:19:15: 70%|███████ | 70/100 [00:00<00:00, 2096.03 it/sec] INFO - 16:19:15: 71%|███████ | 71/100 [00:00<00:00, 2094.60 it/sec] INFO - 16:19:15: 72%|███████▏ | 72/100 [00:00<00:00, 2096.34 it/sec] INFO - 16:19:15: 73%|███████▎ | 73/100 [00:00<00:00, 2096.33 it/sec] INFO - 16:19:15: 74%|███████▍ | 74/100 [00:00<00:00, 2096.36 it/sec] INFO - 16:19:15: 75%|███████▌ | 75/100 [00:00<00:00, 2097.10 it/sec] INFO - 16:19:15: 76%|███████▌ | 76/100 [00:00<00:00, 2096.75 it/sec] INFO - 16:19:15: 77%|███████▋ | 77/100 [00:00<00:00, 2096.91 it/sec] INFO - 16:19:15: 78%|███████▊ | 78/100 [00:00<00:00, 2093.34 it/sec] INFO - 16:19:15: 79%|███████▉ | 79/100 [00:00<00:00, 2086.99 it/sec] INFO - 16:19:15: 80%|████████ | 80/100 [00:00<00:00, 2085.36 it/sec] INFO - 16:19:15: 81%|████████ | 81/100 [00:00<00:00, 2084.61 it/sec] INFO - 16:19:15: 82%|████████▏ | 82/100 [00:00<00:00, 2085.63 it/sec] INFO - 16:19:15: 83%|████████▎ | 83/100 [00:00<00:00, 2085.76 it/sec] INFO - 16:19:15: 84%|████████▍ | 84/100 [00:00<00:00, 2084.84 it/sec] INFO - 16:19:15: 85%|████████▌ | 85/100 [00:00<00:00, 2084.84 it/sec] INFO - 16:19:15: 86%|████████▌ | 86/100 [00:00<00:00, 2084.13 it/sec] INFO - 16:19:15: 87%|████████▋ | 87/100 [00:00<00:00, 2084.60 it/sec] INFO - 16:19:15: 88%|████████▊ | 88/100 [00:00<00:00, 2084.89 it/sec] INFO - 16:19:15: 89%|████████▉ | 89/100 [00:00<00:00, 2084.62 it/sec] INFO - 16:19:15: 90%|█████████ | 90/100 [00:00<00:00, 2085.00 it/sec] INFO - 16:19:15: 91%|█████████ | 91/100 [00:00<00:00, 2085.50 it/sec] INFO - 16:19:15: 92%|█████████▏| 92/100 [00:00<00:00, 2085.07 it/sec] INFO - 16:19:15: 93%|█████████▎| 93/100 [00:00<00:00, 2086.61 it/sec] INFO - 16:19:15: 94%|█████████▍| 94/100 [00:00<00:00, 2085.58 it/sec] INFO - 16:19:15: 95%|█████████▌| 95/100 [00:00<00:00, 2087.17 it/sec] INFO - 16:19:15: 96%|█████████▌| 96/100 [00:00<00:00, 2086.96 it/sec] INFO - 16:19:15: 97%|█████████▋| 97/100 [00:00<00:00, 2088.45 it/sec] INFO - 16:19:15: 98%|█████████▊| 98/100 [00:00<00:00, 2086.97 it/sec] INFO - 16:19:15: 99%|█████████▉| 99/100 [00:00<00:00, 2088.44 it/sec] INFO - 16:19:15: 100%|██████████| 100/100 [00:00<00:00, 2069.54 it/sec] INFO - 16:19:15: *** End Sampling execution *** .. GENERATED FROM PYTHON SOURCE LINES 106-113 Settings -------- Number of estimators ~~~~~~~~~~~~~~~~~~~~ The main hyperparameter of random forest regression is the number of trees in the forest (default: 100). Here is a comparison when increasing and decreasing this number: .. GENERATED FROM PYTHON SOURCE LINES 113-131 .. code-block:: Python model = create_regression_model( "RandomForestRegressor", training_dataset, n_estimators=10 ) model.learn() predicted_output_data_1 = model.predict(input_data).ravel() model = create_regression_model( "RandomForestRegressor", training_dataset, n_estimators=1000 ) model.learn() predicted_output_data_2 = model.predict(input_data).ravel() plt.plot(input_data.ravel(), reference_output_data, label="Reference") plt.plot(input_data.ravel(), predicted_output_data, label="Regression - Basics") plt.plot(input_data.ravel(), predicted_output_data_1, label="Regression - 10 trees") plt.plot(input_data.ravel(), predicted_output_data_2, label="Regression - 1000 trees") plt.grid() plt.legend() plt.show() .. image-sg:: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_002.png :alt: plot random forest regression :srcset: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_002.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 132-142 Others ------ The ``RandomForestRegressor`` class of scikit-learn has a lot of settings (`read more `__), and we have chosen to exhibit only ``n_estimators``. However, any argument of ``RandomForestRegressor`` can be set using the dictionary ``parameters``. For example, we can impose a minimum of two samples per leaf: .. GENERATED FROM PYTHON SOURCE LINES 142-153 .. code-block:: Python model = create_regression_model( "RandomForestRegressor", training_dataset, parameters={"min_samples_leaf": 2} ) model.learn() predicted_output_data_ = model.predict(input_data).ravel() plt.plot(input_data.ravel(), reference_output_data, label="Reference") plt.plot(input_data.ravel(), predicted_output_data, label="Regression - Basics") plt.plot(input_data.ravel(), predicted_output_data_, label="Regression - 2 samples") plt.grid() plt.legend() plt.show() .. image-sg:: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_003.png :alt: plot random forest regression :srcset: /examples/mlearning/regression_model/images/sphx_glr_plot_random_forest_regression_003.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-timing **Total running time of the script:** (0 minutes 1.290 seconds) .. _sphx_glr_download_examples_mlearning_regression_model_plot_random_forest_regression.py: .. only:: html .. container:: sphx-glr-footer sphx-glr-footer-example .. container:: sphx-glr-download sphx-glr-download-jupyter :download:`Download Jupyter notebook: plot_random_forest_regression.ipynb ` .. container:: sphx-glr-download sphx-glr-download-python :download:`Download Python source code: plot_random_forest_regression.py ` .. container:: sphx-glr-download sphx-glr-download-zip :download:`Download zipped: plot_random_forest_regression.zip ` .. only:: html .. rst-class:: sphx-glr-signature `Gallery generated by Sphinx-Gallery `_