{ "cells": [ { "cell_type": "markdown", "id": "8fcab085", "metadata": {}, "source": [ "# Model Optimization\n", "\n", "```{important}\n", "Remember that best is context dependent and relative. The best accuracy might not be the best overall. Automatic optimization can only find the best thing in terms of a single score.\n", "```" ] }, { "cell_type": "code", "execution_count": 1, "id": "fe3434d8", "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import seaborn as sns\n", "import pandas as pd\n", "from sklearn import datasets\n", "from sklearn import cluster\n", "from sklearn import svm, datasets\n", "from sklearn.model_selection import GridSearchCV\n", "from sklearn.model_selection import train_test_split\n", "from sklearn import tree" ] }, { "cell_type": "markdown", "id": "c55d1b5f", "metadata": {}, "source": [ "## Model validation\n", "We will work with the iris data again." ] }, { "cell_type": "code", "execution_count": 2, "id": "b18d9504", "metadata": {}, "outputs": [], "source": [ "iris_df = sns.load_dataset('iris')\n", "\n", "\n", "iris_X = iris_df.drop(columns=['species'])\n", "\n", "iris_y = iris_df['species']" ] }, { "cell_type": "markdown", "id": "d87d1baa", "metadata": {}, "source": [ "We will still use the test train split to keep our test data separate from the data that we use to find our preferred parameters." ] }, { "cell_type": "code", "execution_count": 3, "id": "4286cb6c", "metadata": {}, "outputs": [], "source": [ "iris_X_train, iris_X_test, iris_y_train, iris_y_test = train_test_split(iris_X,iris_y,\n", " random_state=0)" ] }, { "cell_type": "markdown", "id": "546e9015", "metadata": {}, "source": [ "````{margin}\n", "```{admonition} Further Reading\n", "the criteria are discussed [in the mathematical formulation](https://scikit-learn.org/stable/modules/tree.html#mathematical-formulation) of the sklearn documentation\n", "```\n", "````\n", "\n", "Today we will optimize a decision tree over three parameters. One is the criterion, which is how it decides where to create thresholds in parameters. Gini is the default and it computes how concentrated each class is at that node, another is entropy, entropy is, generally how random something is. Intuitively these do similar things, which makes sense because they are two ways to make the same choice, but they have slightly different calculations.\n", "\n", "The other two parameters we have seen some before. Max depth is the height of the tree and min smaples per leaf makes it keeps the leaf sizes small." ] }, { "cell_type": "code", "execution_count": 4, "id": "e833c88b", "metadata": {}, "outputs": [], "source": [ "dt = tree.DecisionTreeClassifier()\n", "params_dt = {'criterion':['gini','entropy'],'max_depth':[2,3,4],\n", " 'min_samples_leaf':list(range(2,20,2))}" ] }, { "cell_type": "markdown", "id": "263fa731", "metadata": {}, "source": [ "We will fit it with default CV settings." ] }, { "cell_type": "code", "execution_count": 5, "id": "86f2f92d", "metadata": {}, "outputs": [], "source": [ "dt_opt = GridSearchCV(dt,params_dt)" ] }, { "cell_type": "markdown", "id": "e35414c6", "metadata": {}, "source": [ "then we can fit" ] }, { "cell_type": "code", "execution_count": 6, "id": "850c78dc", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
GridSearchCV(estimator=DecisionTreeClassifier(),\n",
       "             param_grid={'criterion': ['gini', 'entropy'],\n",
       "                         'max_depth': [2, 3, 4],\n",
       "                         'min_samples_leaf': [2, 4, 6, 8, 10, 12, 14, 16, 18]})
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "GridSearchCV(estimator=DecisionTreeClassifier(),\n", " param_grid={'criterion': ['gini', 'entropy'],\n", " 'max_depth': [2, 3, 4],\n", " 'min_samples_leaf': [2, 4, 6, 8, 10, 12, 14, 16, 18]})" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt.fit(iris_X_train,iris_y_train)" ] }, { "cell_type": "markdown", "id": "76cbcfee", "metadata": {}, "source": [ "we can use ti to get predictions" ] }, { "cell_type": "code", "execution_count": 7, "id": "ec7dce21", "metadata": {}, "outputs": [], "source": [ "y_pred = dt_opt.predict(iris_X_test)" ] }, { "cell_type": "markdown", "id": "407d6dbd", "metadata": {}, "source": [ "we can also score it as regular." ] }, { "cell_type": "code", "execution_count": 8, "id": "129b7ecb", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.9473684210526315" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt.score(iris_X_test,iris_y_test)" ] }, { "cell_type": "markdown", "id": "0bb0084b", "metadata": {}, "source": [ "we can also see the best parameters." ] }, { "cell_type": "code", "execution_count": 9, "id": "3167a801", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'criterion': 'gini', 'max_depth': 4, 'min_samples_leaf': 2}" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt.best_params_" ] }, { "cell_type": "markdown", "id": "375455ad", "metadata": {}, "source": [ "we can look at the overall results this way:" ] }, { "cell_type": "code", "execution_count": 10, "id": "e9b3d386", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
mean_fit_timestd_fit_timemean_score_timestd_score_timeparam_criterionparam_max_depthparam_min_samples_leafparamssplit0_test_scoresplit1_test_scoresplit2_test_scoresplit3_test_scoresplit4_test_scoremean_test_scorestd_test_scorerank_test_score
00.0019780.0003290.0012650.000093gini22{'criterion': 'gini', 'max_depth': 2, 'min_sam...0.9565220.9130431.00.9090910.9545450.946640.0333055
10.0017730.0000140.0012680.000166gini24{'criterion': 'gini', 'max_depth': 2, 'min_sam...0.9565220.9130431.00.9090910.9545450.946640.0333055
20.0017520.0000220.0012000.000043gini26{'criterion': 'gini', 'max_depth': 2, 'min_sam...0.9565220.9130431.00.9090910.9545450.946640.0333055
30.0017470.0000190.0011990.000019gini28{'criterion': 'gini', 'max_depth': 2, 'min_sam...0.9565220.9130431.00.9090910.9545450.946640.0333055
40.0017360.0000300.0011880.000019gini210{'criterion': 'gini', 'max_depth': 2, 'min_sam...0.9565220.9130431.00.9090910.9545450.946640.0333055
\n", "
" ], "text/plain": [ " mean_fit_time std_fit_time mean_score_time std_score_time \\\n", "0 0.001978 0.000329 0.001265 0.000093 \n", "1 0.001773 0.000014 0.001268 0.000166 \n", "2 0.001752 0.000022 0.001200 0.000043 \n", "3 0.001747 0.000019 0.001199 0.000019 \n", "4 0.001736 0.000030 0.001188 0.000019 \n", "\n", " param_criterion param_max_depth param_min_samples_leaf \\\n", "0 gini 2 2 \n", "1 gini 2 4 \n", "2 gini 2 6 \n", "3 gini 2 8 \n", "4 gini 2 10 \n", "\n", " params split0_test_score \\\n", "0 {'criterion': 'gini', 'max_depth': 2, 'min_sam... 0.956522 \n", "1 {'criterion': 'gini', 'max_depth': 2, 'min_sam... 0.956522 \n", "2 {'criterion': 'gini', 'max_depth': 2, 'min_sam... 0.956522 \n", "3 {'criterion': 'gini', 'max_depth': 2, 'min_sam... 0.956522 \n", "4 {'criterion': 'gini', 'max_depth': 2, 'min_sam... 0.956522 \n", "\n", " split1_test_score split2_test_score split3_test_score split4_test_score \\\n", "0 0.913043 1.0 0.909091 0.954545 \n", "1 0.913043 1.0 0.909091 0.954545 \n", "2 0.913043 1.0 0.909091 0.954545 \n", "3 0.913043 1.0 0.909091 0.954545 \n", "4 0.913043 1.0 0.909091 0.954545 \n", "\n", " mean_test_score std_test_score rank_test_score \n", "0 0.94664 0.033305 5 \n", "1 0.94664 0.033305 5 \n", "2 0.94664 0.033305 5 \n", "3 0.94664 0.033305 5 \n", "4 0.94664 0.033305 5 " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_5cv_df = pd.DataFrame(dt_opt.cv_results_)\n", "dt_5cv_df.head()" ] }, { "cell_type": "markdown", "id": "2b97bdf6", "metadata": {}, "source": [ "To see more carefully what it does, we can look at its shape." ] }, { "cell_type": "code", "execution_count": 11, "id": "4ddbe594", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(54, 16)" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_5cv_df.shape" ] }, { "cell_type": "markdown", "id": "ef00e397", "metadata": {}, "source": [ "we can see that the total number of rows matched the product of the length of each list of parameters to try." ] }, { "cell_type": "code", "execution_count": 12, "id": "12c4d5ef", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "54" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "np.product([len(param_list) for param_list in params_dt.values()])" ] }, { "cell_type": "markdown", "id": "320739f3", "metadata": {}, "source": [ "This means that it tests every combination of features-- without us writing a bunch of nested loops.\n", "\n", "We can also plot the dta and look at the performance." ] }, { "cell_type": "code", "execution_count": 13, "id": "0e428a9e", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "filenames": { "image/png": "/home/runner/work/BrownFall22/BrownFall22/_build/jupyter_execute/notes/2022-11-09_25_1.png" } }, "output_type": "display_data" } ], "source": [ "sns.catplot(data=dt_5cv_df,x='param_min_samples_leaf',y='mean_test_score',\n", " col='param_criterion', row= 'param_max_depth', kind='bar',)" ] }, { "cell_type": "markdown", "id": "4cab92ba", "metadata": {}, "source": [ "this makes it clear that none of these stick out much in terms of performance.\n", "\n", "\n", "\n", "## Impact of CV parameters" ] }, { "cell_type": "code", "execution_count": 14, "id": "82994060", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
GridSearchCV(cv=10, estimator=DecisionTreeClassifier(),\n",
       "             param_grid={'criterion': ['gini', 'entropy'],\n",
       "                         'max_depth': [2, 3, 4],\n",
       "                         'min_samples_leaf': [2, 4, 6, 8, 10, 12, 14, 16, 18]})
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "GridSearchCV(cv=10, estimator=DecisionTreeClassifier(),\n", " param_grid={'criterion': ['gini', 'entropy'],\n", " 'max_depth': [2, 3, 4],\n", " 'min_samples_leaf': [2, 4, 6, 8, 10, 12, 14, 16, 18]})" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt10 = GridSearchCV(dt,params_dt,cv=10)\n", "dt_opt10.fit(iris_X_train,iris_y_train)" ] }, { "cell_type": "code", "execution_count": 15, "id": "57342419", "metadata": {}, "outputs": [], "source": [ "dt_10cv_df = pd.DataFrame(dt_opt10.cv_results_)" ] }, { "cell_type": "markdown", "id": "7969f9a9", "metadata": {}, "source": [ "We can stack the columns we want from the two results together with a new indicator column `cv`" ] }, { "cell_type": "code", "execution_count": 16, "id": "2f0d6e28", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
param_min_samples_leafstd_test_scoremean_test_scoreparam_criterionparam_max_depthcv
020.0333050.94664gini25
140.0333050.94664gini25
260.0333050.94664gini25
380.0333050.94664gini25
4100.0333050.94664gini25
\n", "
" ], "text/plain": [ " param_min_samples_leaf std_test_score mean_test_score param_criterion \\\n", "0 2 0.033305 0.94664 gini \n", "1 4 0.033305 0.94664 gini \n", "2 6 0.033305 0.94664 gini \n", "3 8 0.033305 0.94664 gini \n", "4 10 0.033305 0.94664 gini \n", "\n", " param_max_depth cv \n", "0 2 5 \n", "1 2 5 \n", "2 2 5 \n", "3 2 5 \n", "4 2 5 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "plot_cols = ['param_min_samples_leaf','std_test_score','mean_test_score',\n", " 'param_criterion','param_max_depth','cv']\n", "dt_10cv_df['cv'] = 10\n", "dt_5cv_df['cv'] = 5\n", "\n", "dt_cv_df = pd.concat([dt_5cv_df[plot_cols],dt_10cv_df[plot_cols]])\n", "dt_cv_df.head()" ] }, { "cell_type": "markdown", "id": "e5564219", "metadata": {}, "source": [ "this can be used to plot." ] }, { "cell_type": "code", "execution_count": 17, "id": "c9e82635", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "filenames": { "image/png": "/home/runner/work/BrownFall22/BrownFall22/_build/jupyter_execute/notes/2022-11-09_32_1.png" } }, "output_type": "display_data" } ], "source": [ "sns.catplot(data=dt_cv_df,x='param_min_samples_leaf',y='mean_test_score',\n", " col='param_criterion', row= 'param_max_depth', kind='bar',\n", " hue = 'cv')" ] }, { "cell_type": "markdown", "id": "9546a513", "metadata": {}, "source": [ "we see that the mean scores are not very different, but that 10 is a little higher in some cases. This makes sense, it has more data to learn from, so it found something that applied better, on average, to the test set." ] }, { "cell_type": "code", "execution_count": 18, "id": "aec9a1e0", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "filenames": { "image/png": "/home/runner/work/BrownFall22/BrownFall22/_build/jupyter_execute/notes/2022-11-09_34_1.png" } }, "output_type": "display_data" } ], "source": [ "sns.catplot(data=dt_cv_df,x='param_min_samples_leaf',y='std_test_score',\n", " col='param_criterion', row= 'param_max_depth', kind='bar',\n", " hue = 'cv')" ] }, { "cell_type": "markdown", "id": "092ed9e4", "metadata": {}, "source": [ "However here we see that the variabilty in those scores is much higher, so maybe the 5 is better.\n", "\n", "We can compare to see if it finds the same model as best:" ] }, { "cell_type": "code", "execution_count": 19, "id": "e076a61c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'criterion': 'gini', 'max_depth': 4, 'min_samples_leaf': 2}" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt.best_params_" ] }, { "cell_type": "code", "execution_count": 20, "id": "3d07b49f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'criterion': 'gini', 'max_depth': 3, 'min_samples_leaf': 2}" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt10.best_params_" ] }, { "cell_type": "markdown", "id": "6ced0170", "metadata": {}, "source": [ "In some cases they will and others they will not." ] }, { "cell_type": "code", "execution_count": 21, "id": "4575c67c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.9473684210526315" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt.score(iris_X_test,iris_y_test)" ] }, { "cell_type": "code", "execution_count": 22, "id": "61ae3163", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.9736842105263158" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dt_opt10.score(iris_X_test,iris_y_test)" ] }, { "cell_type": "markdown", "id": "d4b65cb6", "metadata": {}, "source": [ "In some cases they will find the same model and score the same, but it other time they will not.\n", "\n", "The takeaway is that the cross validation parameters impact our ability to measure the score and possibly how close that cross validation mean score will match the true test score. Mostly it will change the variability in the estimate of the score. It does not change necessarily which model is best, that is up to the data iteself (the original test/train split would impact this).\n", "\n", "```{admonition} Try it yourself\n", "Does this vary if you repeat this with different test sets? How much does it depend on that? Does repeating it produce the same scores? Does one take longer to fit or score?\n", "```\n", "\n", "\n", "\n", "## Other searches" ] }, { "cell_type": "code", "execution_count": 23, "id": "aad354af", "metadata": {}, "outputs": [], "source": [ "from sklearn import model_selection\n", "from sklearn.model_selection import LeaveOneOut" ] }, { "cell_type": "code", "execution_count": 24, "id": "e58f1113", "metadata": {}, "outputs": [], "source": [ "rand_opt = model_selection.RandomizedSearchCV(dt,params_dt).fit(iris_X_train,\n", " iris_y_train)" ] }, { "cell_type": "code", "execution_count": 25, "id": "9c3012ab", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.9736842105263158" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rand_opt.score(iris_X_test,iris_y_test)" ] }, { "cell_type": "code", "execution_count": 26, "id": "730d448c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'min_samples_leaf': 2, 'max_depth': 3, 'criterion': 'entropy'}" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rand_opt.best_params_" ] }, { "cell_type": "markdown", "id": "4887f9b7", "metadata": {}, "source": [ "It might find the same solution, but it also might not. If you do some and see that the parameters overall do not impact the scores much, then you can trust whichever one, or consider other criteria to choose the best model to use." ] }, { "cell_type": "code", "execution_count": null, "id": "5c30760e", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "26e84828", "metadata": {}, "source": [ "## Questions after class\n", "\n", "### What does cross-validation do that clustering doesn't?\n", "\n", "Cross validation can be used *with* clustering if you want. Cross validation is a technique for evaluating how well a model performs; in this case we are combining with a search to find the best parameters.\n", "\n", "### Is grid search commonly used or is it too expensive for more realistically sized datasets?\n", "\n", "Grid search is realistic for simple models. It becomes more expensive with models that have more options for parameters. For example a parameter that can be continuously valued or take a large range, there become too many to test, so we use the random search or something else. The cost does go up some with larger datasets, but this scales primarily with the nubmer of parameters and values for each.\n", "\n", "For example, bayesian optimization is a more complex type of random search.\n", "\n", "### What is the difference between gini and entropy?\n", "\n", "the criteria are discussed [in the mathematical formulation](https://scikit-learn.org/stable/modules/tree.html#mathematical-formulation) of the sklearn documentation\n", "\n", "### Is it more accurate to do cross vaididation or test_train_split?\n", "\n", "Ideally both: use train test split to set aside a test set for the value that you will report when you say this is how good my final model is and cross validation to choose which model you call the final model." ] } ], "metadata": { "jupytext": { "text_representation": { "extension": ".md", "format_name": "myst", "format_version": 0.13, "jupytext_version": "1.14.1" } }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.16" }, "source_map": [ 12, 20, 31, 36, 43, 47, 50, 62, 66, 70, 72, 76, 78, 82, 84, 88, 90, 94, 96, 100, 103, 106, 108, 111, 113, 119, 122, 130, 135, 137, 140, 148, 151, 155, 159, 163, 169, 173, 175, 179, 183, 185, 199, 204, 209, 213, 215, 219, 221 ] }, "nbformat": 4, "nbformat_minor": 5 }