From 2556da4803138650ff1a6e9e530cdde49798dfd0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Xavier=20Dupr=C3=A9?= Date: Sun, 18 Jan 2026 13:13:00 +0100 Subject: [PATCH 1/2] style --- _doc/conf.py | 1 - _doc/examples/prog/plot_einstein_riddle.py | 2 +- _doc/examples/prog/plot_float_and_double_rouding.py | 1 - _doc/examples/prog/plot_gil_example.py | 1 + _doc/examples/prog/plot_numpy_tricks.py | 1 + _doc/examples/prog/plot_pandas_groupby.py | 2 -- _doc/examples/prog/plot_serialisation_examples.py | 2 +- _doc/examples/prog/plot_serialisation_protobuf.py | 1 + _doc/examples/prog/plot_tarabiscote.py | 1 + _doc/practice/py-base/scrapping.ipynb | 1 - _doc/practice/tds-base/module_file_regex_correction.ipynb | 4 +--- _doc/practice/tds-base/texte_langue_correction.ipynb | 6 ++---- _latex/ensae/td_note_2010_rattrape.py | 6 +----- teachpyx/datasets/wines.py | 1 - teachpyx/tools/display/pygame_helper.py | 1 - 15 files changed, 10 insertions(+), 21 deletions(-) diff --git a/_doc/conf.py b/_doc/conf.py index ba8e12f4..a5f94529 100644 --- a/_doc/conf.py +++ b/_doc/conf.py @@ -4,7 +4,6 @@ from sphinx_runpython.github_link import make_linkcode_resolve from teachpyx import __version__ - extensions = [ "nbsphinx", "sphinx.ext.autodoc", diff --git a/_doc/examples/prog/plot_einstein_riddle.py b/_doc/examples/prog/plot_einstein_riddle.py index da1de5f9..5c4fd03f 100644 --- a/_doc/examples/prog/plot_einstein_riddle.py +++ b/_doc/examples/prog/plot_einstein_riddle.py @@ -87,11 +87,11 @@ On commence par la fonction `permutation`: qui énumère les permutations d'un ensemble : """ + import copy from io import StringIO import pandas - ########################## # Fonction permutation # ==================== diff --git a/_doc/examples/prog/plot_float_and_double_rouding.py b/_doc/examples/prog/plot_float_and_double_rouding.py index d6d09f1f..f1847555 100644 --- a/_doc/examples/prog/plot_float_and_double_rouding.py +++ b/_doc/examples/prog/plot_float_and_double_rouding.py @@ -20,7 +20,6 @@ import pandas import matplotlib.pyplot as plt - rnd = numpy.random.random(100000000) rnd.shape, rnd.dtype diff --git a/_doc/examples/prog/plot_gil_example.py b/_doc/examples/prog/plot_gil_example.py index 4e64c814..bd7ce23a 100644 --- a/_doc/examples/prog/plot_gil_example.py +++ b/_doc/examples/prog/plot_gil_example.py @@ -17,6 +17,7 @@ On mesure le temps nécessaire pour créer deux liste et comparer ce temps avec celui que cela prendrait en parallèle. """ + import timeit import time from concurrent.futures import ThreadPoolExecutor diff --git a/_doc/examples/prog/plot_numpy_tricks.py b/_doc/examples/prog/plot_numpy_tricks.py index 3c732436..92d0b67a 100644 --- a/_doc/examples/prog/plot_numpy_tricks.py +++ b/_doc/examples/prog/plot_numpy_tricks.py @@ -9,6 +9,7 @@ accéder à un élément en particulier =================================== """ + import timeit import numpy diff --git a/_doc/examples/prog/plot_pandas_groupby.py b/_doc/examples/prog/plot_pandas_groupby.py index 07cf7583..238636a8 100644 --- a/_doc/examples/prog/plot_pandas_groupby.py +++ b/_doc/examples/prog/plot_pandas_groupby.py @@ -14,10 +14,8 @@ ============================ """ - import pandas - data = [{"a": 1, "b": 2}, {"a": 10, "b": 20}, {"b": 3}, {"b": 4}] df = pandas.DataFrame(data) df diff --git a/_doc/examples/prog/plot_serialisation_examples.py b/_doc/examples/prog/plot_serialisation_examples.py index 49fc35f5..d943aacf 100644 --- a/_doc/examples/prog/plot_serialisation_examples.py +++ b/_doc/examples/prog/plot_serialisation_examples.py @@ -17,6 +17,7 @@ Ecriture (json) +++++++++++++++ """ + from io import StringIO, BytesIO import timeit import json @@ -25,7 +26,6 @@ import cloudpickle import pickle - data = { "records": [ { diff --git a/_doc/examples/prog/plot_serialisation_protobuf.py b/_doc/examples/prog/plot_serialisation_protobuf.py index cd71f774..1ff3b301 100644 --- a/_doc/examples/prog/plot_serialisation_protobuf.py +++ b/_doc/examples/prog/plot_serialisation_protobuf.py @@ -25,6 +25,7 @@ On récupère l'exemple du `tutorial `_. """ + import os import sys import timeit diff --git a/_doc/examples/prog/plot_tarabiscote.py b/_doc/examples/prog/plot_tarabiscote.py index a68fbaba..5fc3ac47 100644 --- a/_doc/examples/prog/plot_tarabiscote.py +++ b/_doc/examples/prog/plot_tarabiscote.py @@ -16,6 +16,7 @@ listes contenues dans ``ens``. Le résultat retourné est effectivement celui désiré mais la fonction modifie également la liste ``ens``, pourquoi ? """ + import math import copy import numpy diff --git a/_doc/practice/py-base/scrapping.ipynb b/_doc/practice/py-base/scrapping.ipynb index 18cba6f8..006ee7c6 100644 --- a/_doc/practice/py-base/scrapping.ipynb +++ b/_doc/practice/py-base/scrapping.ipynb @@ -380,7 +380,6 @@ "import shutil\n", "import requests\n", "\n", - "\n", "for e, pokemon in enumerate(liste_pokemon):\n", " print(e, pokemon)\n", " url = \"https://img.pokemondb.net/artwork/{}.jpg\".format(pokemon)\n", diff --git a/_doc/practice/tds-base/module_file_regex_correction.ipynb b/_doc/practice/tds-base/module_file_regex_correction.ipynb index 3dfb03cf..f858e6db 100644 --- a/_doc/practice/tds-base/module_file_regex_correction.ipynb +++ b/_doc/practice/tds-base/module_file_regex_correction.ipynb @@ -275,9 +275,7 @@ " return [ math.cos(x) for x in seq ] \n", " if True :\n", " print (\"Ce message n'apparaît que si ce programme est le point d'entrée.\")\n", - " \"\"\".replace(\n", - " \" \", \"\"\n", - " )\n", + " \"\"\".replace(\" \", \"\")\n", " with open(\"monmodule3.py\", \"w\", encoding=\"utf8\") as f:\n", " f.write(code)" ] diff --git a/_doc/practice/tds-base/texte_langue_correction.ipynb b/_doc/practice/tds-base/texte_langue_correction.ipynb index 095216ce..87a56d64 100644 --- a/_doc/practice/tds-base/texte_langue_correction.ipynb +++ b/_doc/practice/tds-base/texte_langue_correction.ipynb @@ -82,10 +82,8 @@ "outputs": [], "source": [ "with open(\"texte.txt\", \"w\", encoding=\"utf-8\") as f:\n", - " f.write(\n", - " \"\"\"Un corbeau sur un arbre perché tenait en son bec un fromage.\n", - "Maître Renard, par l'odeur alléché, Lui tint à peu près ce langage :\"\"\"\n", - " )" + " f.write(\"\"\"Un corbeau sur un arbre perché tenait en son bec un fromage.\n", + "Maître Renard, par l'odeur alléché, Lui tint à peu près ce langage :\"\"\")" ] }, { diff --git a/_latex/ensae/td_note_2010_rattrape.py b/_latex/ensae/td_note_2010_rattrape.py index acb0016c..1aad043f 100644 --- a/_latex/ensae/td_note_2010_rattrape.py +++ b/_latex/ensae/td_note_2010_rattrape.py @@ -33,11 +33,7 @@ def get_tour(): Metz 6,11729002 49,0734787 Sedan 4,896070004 49,68407059 Grenoble 5,684440136 45,13940048 -Annecy 6,082499981 45,8782196""".replace( - ",", "." - ).split( - "\n" - ) +Annecy 6,082499981 45,8782196""".replace(",", ".").split("\n") # ligne d'avant : on d�coupe l'unique cha�ne de caract�res # ligne suivant : on d�coupe chaque ligne en colonne diff --git a/teachpyx/datasets/wines.py b/teachpyx/datasets/wines.py index e7b534df..eb4caf46 100644 --- a/teachpyx/datasets/wines.py +++ b/teachpyx/datasets/wines.py @@ -3,7 +3,6 @@ import pandas from .data_helper import get_data_folder - __all__ = ["load_wines_dataset"] diff --git a/teachpyx/tools/display/pygame_helper.py b/teachpyx/tools/display/pygame_helper.py index 56142d34..3399e28b 100644 --- a/teachpyx/tools/display/pygame_helper.py +++ b/teachpyx/tools/display/pygame_helper.py @@ -1,7 +1,6 @@ import math from typing import List, Optional, Tuple - MOUSE = "mouse" KEY = "key" From 98730d91503cd291430253cef5c4f8da43d9961b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Xavier=20Dupr=C3=A9?= Date: Sun, 18 Jan 2026 13:47:37 +0100 Subject: [PATCH 2/2] fix two notebooks --- _doc/practice/ml/winesc_multi.ipynb | 7536 +++++++++++++++--- _doc/practice/ml/winesc_multi_stacking.ipynb | 3391 +++++++- 2 files changed, 9779 insertions(+), 1148 deletions(-) diff --git a/_doc/practice/ml/winesc_multi.ipynb b/_doc/practice/ml/winesc_multi.ipynb index 7975365d..e6a7a087 100644 --- a/_doc/practice/ml/winesc_multi.ipynb +++ b/_doc/practice/ml/winesc_multi.ipynb @@ -11,7 +11,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ @@ -20,7 +20,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 12, "metadata": { "scrolled": true }, @@ -35,7 +35,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 13, "metadata": {}, "outputs": [], "source": [ @@ -46,15 +46,31 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 14, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n" + ] + }, { "data": { "text/html": [ - "
LogisticRegression(solver='liblinear')
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCA0NDggNTEyIj48IS0tIUZvbnQgQXdlc29tZSBGcmVlIDYuNy4yIGJ5IEBmb250YXdlc29tZSAtIGh0dHBzOi8vZm9udGF3ZXNvbWUuY29tIExpY2Vuc2UgLSBodHRwczovL2ZvbnRhd2Vzb21lLmNvbS9saWNlbnNlL2ZyZWUgQ29weXJpZ2h0IDIwMjUgRm9udGljb25zLCBJbmMuLS0+PHBhdGggZD0iTTIwOCAwTDMzMi4xIDBjMTIuNyAwIDI0LjkgNS4xIDMzLjkgMTQuMWw2Ny45IDY3LjljOSA5IDE0LjEgMjEuMiAxNC4xIDMzLjlMNDQ4IDMzNmMwIDI2LjUtMjEuNSA0OC00OCA0OGwtMTkyIDBjLTI2LjUgMC00OC0yMS41LTQ4LTQ4bDAtMjg4YzAtMjYuNSAyMS41LTQ4IDQ4LTQ4ek00OCAxMjhsODAgMCAwIDY0LTY0IDAgMCAyNTYgMTkyIDAgMC0zMiA2NCAwIDAgNDhjMCAyNi41LTIxLjUgNDgtNDggNDhMNDggNTEyYy0yNi41IDAtNDgtMjEuNS00OC00OEwwIDE3NmMwLTI2LjUgMjEuNS00OCA0OC00OHoiLz48L3N2Zz4=);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
LogisticRegression()
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": [ - "LogisticRegression(solver='liblinear')" + "LogisticRegression()" ] }, - "execution_count": 6, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" } @@ -470,22 +980,30 @@ "source": [ "from sklearn.linear_model import LogisticRegression\n", "\n", - "clr = LogisticRegression(solver=\"liblinear\")\n", + "clr = LogisticRegression()\n", "clr.fit(X_train, y_train)" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 15, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/2645516986.py:3: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "55.07692307692308" + "np.float64(46.52307692307692)" ] }, - "execution_count": 7, + "execution_count": 15, "metadata": {}, "output_type": "execute_result" } @@ -505,7 +1023,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 16, "metadata": {}, "outputs": [ { @@ -535,7 +1053,6 @@ " 3\n", " 4\n", " 5\n", - " 6\n", " \n", " \n", " \n", @@ -543,9 +1060,8 @@ " 0\n", " 0\n", " 0\n", - " 4\n", - " 7\n", - " 0\n", + " 5\n", + " 5\n", " 0\n", " 0\n", " \n", @@ -553,60 +1069,45 @@ " 1\n", " 0\n", " 0\n", - " 47\n", - " 22\n", - " 0\n", - " 0\n", + " 15\n", + " 39\n", + " 1\n", " 0\n", " \n", " \n", " 2\n", " 0\n", " 0\n", - " 332\n", - " 184\n", - " 0\n", - " 0\n", + " 240\n", + " 316\n", + " 5\n", " 0\n", " \n", " \n", " 3\n", " 0\n", " 0\n", - " 169\n", - " 541\n", - " 10\n", - " 0\n", + " 177\n", + " 511\n", + " 7\n", " 0\n", " \n", " \n", " 4\n", " 0\n", " 0\n", - " 19\n", - " 217\n", - " 22\n", - " 0\n", - " 0\n", - " \n", - " \n", - " 5\n", - " 0\n", - " 0\n", - " 3\n", - " 42\n", + " 35\n", + " 212\n", " 5\n", " 0\n", - " 0\n", " \n", " \n", - " 6\n", - " 0\n", - " 0\n", - " 0\n", - " 1\n", + " 5\n", " 0\n", " 0\n", + " 6\n", + " 44\n", + " 2\n", " 0\n", " \n", " \n", @@ -614,17 +1115,16 @@ "" ], "text/plain": [ - " 0 1 2 3 4 5 6\n", - "0 0 0 4 7 0 0 0\n", - "1 0 0 47 22 0 0 0\n", - "2 0 0 332 184 0 0 0\n", - "3 0 0 169 541 10 0 0\n", - "4 0 0 19 217 22 0 0\n", - "5 0 0 3 42 5 0 0\n", - "6 0 0 0 1 0 0 0" + " 0 1 2 3 4 5\n", + "0 0 0 5 5 0 0\n", + "1 0 0 15 39 1 0\n", + "2 0 0 240 316 5 0\n", + "3 0 0 177 511 7 0\n", + "4 0 0 35 212 5 0\n", + "5 0 0 6 44 2 0" ] }, - "execution_count": 8, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } @@ -645,7 +1145,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 17, "metadata": {}, "outputs": [ { @@ -675,7 +1175,6 @@ " 6\n", " 7\n", " 8\n", - " 9\n", " \n", " \n", " \n", @@ -683,9 +1182,8 @@ " 3\n", " 0\n", " 0\n", - " 4\n", - " 7\n", - " 0\n", + " 5\n", + " 5\n", " 0\n", " 0\n", " \n", @@ -693,60 +1191,45 @@ " 4\n", " 0\n", " 0\n", - " 47\n", - " 22\n", - " 0\n", - " 0\n", + " 15\n", + " 39\n", + " 1\n", " 0\n", " \n", " \n", " 5\n", " 0\n", " 0\n", - " 332\n", - " 184\n", - " 0\n", - " 0\n", + " 240\n", + " 316\n", + " 5\n", " 0\n", " \n", " \n", " 6\n", " 0\n", " 0\n", - " 169\n", - " 541\n", - " 10\n", - " 0\n", + " 177\n", + " 511\n", + " 7\n", " 0\n", " \n", " \n", " 7\n", " 0\n", " 0\n", - " 19\n", - " 217\n", - " 22\n", - " 0\n", - " 0\n", - " \n", - " \n", - " 8\n", - " 0\n", - " 0\n", - " 3\n", - " 42\n", + " 35\n", + " 212\n", " 5\n", " 0\n", - " 0\n", " \n", " \n", - " 9\n", - " 0\n", - " 0\n", - " 0\n", - " 1\n", + " 8\n", " 0\n", " 0\n", + " 6\n", + " 44\n", + " 2\n", " 0\n", " \n", " \n", @@ -754,17 +1237,16 @@ "" ], "text/plain": [ - " 3 4 5 6 7 8 9\n", - "3 0 0 4 7 0 0 0\n", - "4 0 0 47 22 0 0 0\n", - "5 0 0 332 184 0 0 0\n", - "6 0 0 169 541 10 0 0\n", - "7 0 0 19 217 22 0 0\n", - "8 0 0 3 42 5 0 0\n", - "9 0 0 0 1 0 0 0" + " 3 4 5 6 7 8\n", + "3 0 0 5 5 0 0\n", + "4 0 0 15 39 1 0\n", + "5 0 0 240 316 5 0\n", + "6 0 0 177 511 7 0\n", + "7 0 0 35 212 5 0\n", + "8 0 0 6 44 2 0" ] }, - "execution_count": 9, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -793,15 +1275,16 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsRestClassifier(estimator=LogisticRegression(solver='liblinear'))
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": [ - "OneVsRestClassifier(estimator=LogisticRegression(solver='liblinear'))" - ] - }, - "execution_count": 10, - "metadata": {}, - "output_type": "execute_result" - } - ], + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsRestClassifier(estimator=LogisticRegression(solver='liblinear'))
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": [ + "OneVsRestClassifier(estimator=LogisticRegression(solver='liblinear'))" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], "source": [ "from sklearn.multiclass import OneVsRestClassifier\n", "\n", @@ -1223,16 +2258,24 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 19, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "54.95384615384615" + "np.float64(52.184615384615384)" ] }, - "execution_count": 11, + "execution_count": 19, "metadata": {}, "output_type": "execute_result" } @@ -1250,15 +2293,16 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsOneClassifier(estimator=LogisticRegression(solver='liblinear'))
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsOneClassifier(estimator=LogisticRegression(solver='liblinear'))
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": [ "OneVsOneClassifier(estimator=LogisticRegression(solver='liblinear'))" ] }, - "execution_count": 12, + "execution_count": 20, "metadata": {}, "output_type": "execute_result" } @@ -1680,16 +3260,24 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 21, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "55.138461538461534" + "np.float64(52.12307692307693)" ] }, - "execution_count": 13, + "execution_count": 21, "metadata": {}, "output_type": "execute_result" } @@ -1700,7 +3288,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 22, "metadata": {}, "outputs": [ { @@ -1730,7 +3318,6 @@ " 6\n", " 7\n", " 8\n", - " 9\n", " \n", " \n", " \n", @@ -1738,19 +3325,17 @@ " 3\n", " 0\n", " 0\n", - " 5\n", + " 4\n", " 6\n", " 0\n", " 0\n", - " 0\n", " \n", " \n", " 4\n", " 0\n", - " 0\n", - " 46\n", - " 23\n", - " 0\n", + " 1\n", + " 27\n", + " 27\n", " 0\n", " 0\n", " \n", @@ -1758,20 +3343,18 @@ " 5\n", " 0\n", " 0\n", - " 332\n", - " 183\n", + " 329\n", + " 229\n", + " 2\n", " 1\n", - " 0\n", - " 0\n", " \n", " \n", " 6\n", " 0\n", " 0\n", - " 169\n", - " 524\n", - " 27\n", - " 0\n", + " 172\n", + " 485\n", + " 38\n", " 0\n", " \n", " \n", @@ -1779,29 +3362,17 @@ " 0\n", " 0\n", " 18\n", - " 200\n", - " 40\n", - " 0\n", - " 0\n", - " \n", - " \n", - " 8\n", - " 0\n", - " 0\n", - " 6\n", + " 202\n", " 32\n", - " 12\n", - " 0\n", " 0\n", " \n", " \n", - " 9\n", - " 0\n", - " 0\n", - " 0\n", - " 1\n", + " 8\n", " 0\n", " 0\n", + " 5\n", + " 30\n", + " 17\n", " 0\n", " \n", " \n", @@ -1809,17 +3380,16 @@ "" ], "text/plain": [ - " 3 4 5 6 7 8 9\n", - "3 0 0 5 6 0 0 0\n", - "4 0 0 46 23 0 0 0\n", - "5 0 0 332 183 1 0 0\n", - "6 0 0 169 524 27 0 0\n", - "7 0 0 18 200 40 0 0\n", - "8 0 0 6 32 12 0 0\n", - "9 0 0 0 1 0 0 0" + " 3 4 5 6 7 8\n", + "3 0 0 4 6 0 0\n", + "4 0 1 27 27 0 0\n", + "5 0 0 329 229 2 1\n", + "6 0 0 172 485 38 0\n", + "7 0 0 18 202 32 0\n", + "8 0 0 5 30 17 0" ] }, - "execution_count": 14, + "execution_count": 22, "metadata": {}, "output_type": "execute_result" } @@ -1848,15 +3418,16 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
DecisionTreeClassifier()
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
DecisionTreeClassifier()
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": [ "DecisionTreeClassifier()" ] }, - "execution_count": 15, + "execution_count": 23, "metadata": {}, "output_type": "execute_result" } @@ -2278,16 +4327,24 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 24, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "59.323076923076925" + "np.float64(59.50769230769231)" ] }, - "execution_count": 16, + "execution_count": 24, "metadata": {}, "output_type": "execute_result" } @@ -2305,15 +4362,16 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsRestClassifier(estimator=DecisionTreeClassifier())
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsRestClassifier(estimator=DecisionTreeClassifier())
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": [ "OneVsRestClassifier(estimator=DecisionTreeClassifier())" ] }, - "execution_count": 17, + "execution_count": 25, "metadata": {}, "output_type": "execute_result" } @@ -2733,16 +5327,24 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 26, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "53.35384615384615" + "np.float64(56.61538461538461)" ] }, - "execution_count": 18, + "execution_count": 26, "metadata": {}, "output_type": "execute_result" } @@ -2760,15 +5362,16 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsOneClassifier(estimator=DecisionTreeClassifier())
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsOneClassifier(estimator=DecisionTreeClassifier())
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": [ "OneVsOneClassifier(estimator=DecisionTreeClassifier())" ] }, - "execution_count": 19, + "execution_count": 27, "metadata": {}, "output_type": "execute_result" } @@ -3188,16 +6311,24 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 28, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "62.58461538461538" + "np.float64(59.815384615384616)" ] }, - "execution_count": 20, + "execution_count": 28, "metadata": {}, "output_type": "execute_result" } @@ -3215,15 +6346,16 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
RandomForestClassifier()
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
RandomForestClassifier()
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": [ "RandomForestClassifier()" ] }, - "execution_count": 21, + "execution_count": 29, "metadata": {}, "output_type": "execute_result" } @@ -3645,16 +7351,24 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 30, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "69.2923076923077" + "np.float64(67.44615384615385)" ] }, - "execution_count": 23, + "execution_count": 30, "metadata": {}, "output_type": "execute_result" } @@ -3665,15 +7379,16 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsRestClassifier(estimator=RandomForestClassifier())
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": [ "OneVsRestClassifier(estimator=RandomForestClassifier())" ] }, - "execution_count": 24, + "execution_count": 31, "metadata": {}, "output_type": "execute_result" } @@ -4093,16 +8440,24 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 32, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "69.41538461538461" + "np.float64(67.75384615384615)" ] }, - "execution_count": 25, + "execution_count": 32, "metadata": {}, "output_type": "execute_result" } @@ -4120,15 +8475,16 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
MLPClassifier(hidden_layer_sizes=30, max_iter=600)
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
MLPClassifier(hidden_layer_sizes=30, max_iter=600)
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": [ "MLPClassifier(hidden_layer_sizes=30, max_iter=600)" ] }, - "execution_count": 26, + "execution_count": 33, "metadata": {}, "output_type": "execute_result" } @@ -4550,16 +9544,24 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 34, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "52.800000000000004" + "np.float64(51.38461538461539)" ] }, - "execution_count": 27, + "execution_count": 34, "metadata": {}, "output_type": "execute_result" } @@ -4570,15 +9572,16 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
OneVsRestClassifier(estimator=MLPClassifier(hidden_layer_sizes=30,\n",
-       "                                            max_iter=600))
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsRestClassifier(estimator=MLPClassifier(hidden_layer_sizes=30,\n",
+       "                                            max_iter=600))
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": [ "OneVsRestClassifier(estimator=MLPClassifier(hidden_layer_sizes=30,\n", " max_iter=600))" ] }, - "execution_count": 28, + "execution_count": 35, "metadata": {}, "output_type": "execute_result" } @@ -5001,16 +10699,24 @@ }, { "cell_type": "code", - "execution_count": 29, + "execution_count": 36, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_271755/1706611708.py:1: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "52.800000000000004" + "np.float64(51.815384615384616)" ] }, - "execution_count": 29, + "execution_count": 36, "metadata": {}, "output_type": "execute_result" } @@ -5029,7 +10735,7 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3 (ipykernel)", + "display_name": "this312", "language": "python", "name": "python3" }, @@ -5043,7 +10749,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.12" + "version": "3.12.3" } }, "nbformat": 4, diff --git a/_doc/practice/ml/winesc_multi_stacking.ipynb b/_doc/practice/ml/winesc_multi_stacking.ipynb index 7b94f5a3..9daa5100 100644 --- a/_doc/practice/ml/winesc_multi_stacking.ipynb +++ b/_doc/practice/ml/winesc_multi_stacking.ipynb @@ -20,7 +20,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 2, "metadata": {}, "outputs": [ { @@ -30,7 +30,7 @@ "" ] }, - "execution_count": 1, + "execution_count": 2, "metadata": { "image/png": { "width": 400 @@ -47,7 +47,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ @@ -56,7 +56,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 4, "metadata": { "scrolled": true }, @@ -90,7 +90,8 @@ "text/html": [ "
OneVsRestClassifier(estimator=LogisticRegression(max_iter=1500))
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
OneVsRestClassifier(estimator=LogisticRegression(max_iter=1500))
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": [ "OneVsRestClassifier(estimator=LogisticRegression(max_iter=1500))" @@ -516,10 +1069,18 @@ "execution_count": 7, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_272079/2645516986.py:3: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(clr.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "54.70769230769231" + "np.float64(54.21538461538462)" ] }, "execution_count": 7, @@ -572,7 +1133,6 @@ " 6\n", " 7\n", " 8\n", - " 9\n", " \n", " \n", " \n", @@ -580,70 +1140,54 @@ " 3\n", " 0\n", " 0\n", - " 3\n", - " 5\n", - " 0\n", " 1\n", + " 4\n", + " 0\n", " 0\n", " \n", " \n", " 4\n", " 0\n", " 0\n", - " 40\n", - " 18\n", - " 0\n", - " 0\n", + " 28\n", + " 22\n", + " 1\n", " 0\n", " \n", " \n", " 5\n", " 0\n", " 0\n", - " 339\n", - " 198\n", - " 0\n", - " 0\n", + " 343\n", + " 194\n", + " 2\n", " 0\n", " \n", " \n", " 6\n", " 0\n", " 0\n", - " 184\n", - " 527\n", - " 8\n", - " 0\n", - " 0\n", + " 171\n", + " 515\n", + " 16\n", + " 1\n", " \n", " \n", " 7\n", " 0\n", " 0\n", " 23\n", - " 210\n", + " 224\n", " 23\n", " 0\n", - " 0\n", " \n", " \n", " 8\n", " 0\n", " 0\n", - " 3\n", - " 33\n", - " 9\n", - " 0\n", - " 0\n", - " \n", - " \n", - " 9\n", - " 0\n", - " 0\n", - " 0\n", - " 1\n", - " 0\n", - " 0\n", + " 6\n", + " 49\n", + " 2\n", " 0\n", " \n", " \n", @@ -651,14 +1195,13 @@ "" ], "text/plain": [ - " 3 4 5 6 7 8 9\n", - "3 0 0 3 5 0 1 0\n", - "4 0 0 40 18 0 0 0\n", - "5 0 0 339 198 0 0 0\n", - "6 0 0 184 527 8 0 0\n", - "7 0 0 23 210 23 0 0\n", - "8 0 0 3 33 9 0 0\n", - "9 0 0 0 1 0 0 0" + " 3 4 5 6 7 8\n", + "3 0 0 1 4 0 0\n", + "4 0 0 28 22 1 0\n", + "5 0 0 343 194 2 0\n", + "6 0 0 171 515 16 1\n", + "7 0 0 23 224 23 0\n", + "8 0 0 6 49 2 0" ] }, "execution_count": 8, @@ -693,10 +1236,18 @@ "execution_count": 9, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_272079/717454980.py:5: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(rfc.predict(X_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "68.9846153846154" + "np.float64(69.72307692307692)" ] }, "execution_count": 9, @@ -731,7 +1282,8 @@ "text/html": [ "
RandomForestClassifier()
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
RandomForestClassifier()
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": [ "RandomForestClassifier()" @@ -1163,10 +2289,18 @@ "execution_count": 11, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_272079/3402916033.py:2: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " numpy.mean(rfc_y.predict(rf_test).ravel() == y_test.ravel()) * 100\n" + ] + }, { "data": { "text/plain": [ - "66.83076923076922" + "np.float64(66.70769230769231)" ] }, "execution_count": 11, @@ -1194,7 +2328,7 @@ { "data": { "text/plain": [ - "(0.7269546464802059, 0.6626763197852255, 0.6701197104064432)" + "(0.6219199242594116, 0.5861175474001545, 0.5618693474848643)" ] }, "execution_count": 12, @@ -1223,7 +2357,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -1283,7 +2417,7 @@ { "data": { "text/plain": [ - "(0.5556820682740744, 0.7318154142418192, 0.7349919877544187)" + "(0.5863418891045122, 0.744300691021032, 0.7371402896099202)" ] }, "execution_count": 15, @@ -1326,7 +2460,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -1364,7 +2498,8 @@ "text/html": [ "
RandomForestClassifier()
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
RandomForestClassifier()
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": [ "RandomForestClassifier()" @@ -1799,9 +3508,9 @@ { "data": { "text/plain": [ - "array([[0.42, 0.58],\n", - " [0.89, 0.11],\n", - " [0.64, 0.36]])" + "array([[0.45, 0.55],\n", + " [0.6 , 0.4 ],\n", + " [0.58, 0.42]])" ] }, "execution_count": 18, @@ -1831,7 +3540,7 @@ { "data": { "text/plain": [ - "(0.5556820682740744, 0.7731918190932655)" + "(0.5863418891045122, 0.7804311144471703)" ] }, "execution_count": 20, @@ -1854,7 +3563,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -1893,25 +3602,76 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": 23, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "/home/xadupre/install/scikit-learn/sklearn/model_selection/_split.py:737: UserWarning: The least populated class in y has only 4 members, which is less than n_splits=5.\n", - " warnings.warn(\n", - "/home/xadupre/install/scikit-learn/sklearn/model_selection/_split.py:737: UserWarning: The least populated class in y has only 4 members, which is less than n_splits=5.\n", - " warnings.warn(\n" + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n", + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n", + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n", + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n", + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n", + "/home/xadupre/vv/this312/lib/python3.12/site-packages/sklearn/linear_model/_logistic.py:406: ConvergenceWarning: lbfgs failed to converge after 100 iteration(s) (status=1):\n", + "STOP: TOTAL NO. OF ITERATIONS REACHED LIMIT\n", + "\n", + "Increase the number of iterations to improve the convergence (max_iter=100).\n", + "You might also want to scale the data as shown in:\n", + " https://scikit-learn.org/stable/modules/preprocessing.html\n", + "Please also refer to the documentation for alternative solver options:\n", + " https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", + " n_iter_i = _check_optimize_result(\n" ] }, { "data": { "text/html": [ - "
StackingClassifier(estimators=[('ovrlr',\n",
-       "                                LogisticRegression(solver='liblinear')),\n",
-       "                               ('rf', RandomForestClassifier())])
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.
" + "\n", + ".estimator-table {\n", + " font-family: monospace;\n", + "}\n", + "\n", + ".estimator-table summary {\n", + " padding: .5rem;\n", + " cursor: pointer;\n", + "}\n", + "\n", + ".estimator-table summary::marker {\n", + " font-size: 0.7rem;\n", + "}\n", + "\n", + ".estimator-table details[open] {\n", + " padding-left: 0.1rem;\n", + " padding-right: 0.1rem;\n", + " padding-bottom: 0.3rem;\n", + "}\n", + "\n", + ".estimator-table .parameters-table {\n", + " margin-left: auto !important;\n", + " margin-right: auto !important;\n", + " margin-top: 0;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(odd) {\n", + " background-color: #fff;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:nth-child(even) {\n", + " background-color: #f6f6f6;\n", + "}\n", + "\n", + ".estimator-table .parameters-table tr:hover {\n", + " background-color: #e0e0e0;\n", + "}\n", + "\n", + ".estimator-table table td {\n", + " border: 1px solid rgba(106, 105, 104, 0.232);\n", + "}\n", + "\n", + "/*\n", + " `table td`is set in notebook with right text-align.\n", + " We need to overwrite it.\n", + "*/\n", + ".estimator-table table td.param {\n", + " text-align: left;\n", + " position: relative;\n", + " padding: 0;\n", + "}\n", + "\n", + ".user-set td {\n", + " color:rgb(255, 94, 0);\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td.value {\n", + " color:rgb(255, 94, 0);\n", + " background-color: transparent;\n", + "}\n", + "\n", + ".default td {\n", + " color: black;\n", + " text-align: left !important;\n", + "}\n", + "\n", + ".user-set td i,\n", + ".default td i {\n", + " color: black;\n", + "}\n", + "\n", + "/*\n", + " Styles for parameter documentation links\n", + " We need styling for visited so jupyter doesn't overwrite it\n", + "*/\n", + "a.param-doc-link,\n", + "a.param-doc-link:link,\n", + "a.param-doc-link:visited {\n", + " text-decoration: underline dashed;\n", + " text-underline-offset: .3em;\n", + " color: inherit;\n", + " display: block;\n", + " padding: .5em;\n", + "}\n", + "\n", + "/* \"hack\" to make the entire area of the cell containing the link clickable */\n", + "a.param-doc-link::before {\n", + " position: absolute;\n", + " content: \"\";\n", + " inset: 0;\n", + "}\n", + "\n", + ".param-doc-description {\n", + " display: none;\n", + " position: absolute;\n", + " z-index: 9999;\n", + " left: 0;\n", + " padding: .5ex;\n", + " margin-left: 1.5em;\n", + " color: var(--sklearn-color-text);\n", + " box-shadow: .3em .3em .4em #999;\n", + " width: max-content;\n", + " text-align: left;\n", + " max-height: 10em;\n", + " overflow-y: auto;\n", + "\n", + " /* unfitted */\n", + " background: var(--sklearn-color-unfitted-level-0);\n", + " border: thin solid var(--sklearn-color-unfitted-level-3);\n", + "}\n", + "\n", + "/* Fitted state for parameter tooltips */\n", + ".fitted .param-doc-description {\n", + " /* fitted */\n", + " background: var(--sklearn-color-fitted-level-0);\n", + " border: thin solid var(--sklearn-color-fitted-level-3);\n", + "}\n", + "\n", + ".param-doc-link:hover .param-doc-description {\n", + " display: block;\n", + "}\n", + "\n", + ".copy-paste-icon {\n", + " background-image: url(data:image/svg+xml;base64,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);\n", + " background-repeat: no-repeat;\n", + " background-size: 14px 14px;\n", + " background-position: 0;\n", + " display: inline-block;\n", + " width: 14px;\n", + " height: 14px;\n", + " cursor: pointer;\n", + "}\n", + "
StackingClassifier(estimators=[('ovrlr', LogisticRegression()),\n",
+       "                               ('rf', RandomForestClassifier())])
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": [ - "StackingClassifier(estimators=[('ovrlr',\n", - " LogisticRegression(solver='liblinear')),\n", + "StackingClassifier(estimators=[('ovrlr', LogisticRegression()),\n", " ('rf', RandomForestClassifier())])" ] }, - "execution_count": 31, + "execution_count": 23, "metadata": {}, "output_type": "execute_result" } @@ -2335,7 +5255,7 @@ "\n", "model = StackingClassifier(\n", " [\n", - " (\"ovrlr\", LogisticRegression(solver=\"liblinear\")),\n", + " (\"ovrlr\", LogisticRegression()),\n", " (\"rf\", RandomForestClassifier()),\n", " ]\n", ")\n", @@ -2344,16 +5264,16 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "0.7196235911146174" + "0.7428493449781659" ] }, - "execution_count": 32, + "execution_count": 24, "metadata": {}, "output_type": "execute_result" } @@ -2376,11 +5296,16 @@ "source": [ "La validation croisée a été escamotée par gain de temps mais faire l'impasse est risquée dans le cas général." ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [] } ], "metadata": { "kernelspec": { - "display_name": "Python 3 (ipykernel)", + "display_name": "this312", "language": "python", "name": "python3" }, @@ -2394,7 +5319,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.12" + "version": "3.12.3" } }, "nbformat": 4,