|
a |
|
b/.ipynb_checkpoints/Random Model Stacking-checkpoint.ipynb |
|
|
1 |
{ |
|
|
2 |
"cells": [ |
|
|
3 |
{ |
|
|
4 |
"cell_type": "code", |
|
|
5 |
"execution_count": 16, |
|
|
6 |
"metadata": { |
|
|
7 |
"scrolled": true |
|
|
8 |
}, |
|
|
9 |
"outputs": [ |
|
|
10 |
{ |
|
|
11 |
"name": "stdout", |
|
|
12 |
"output_type": "stream", |
|
|
13 |
"text": [ |
|
|
14 |
"CPU times: user 42 µs, sys: 10 µs, total: 52 µs\n", |
|
|
15 |
"Wall time: 56.5 µs\n" |
|
|
16 |
] |
|
|
17 |
} |
|
|
18 |
], |
|
|
19 |
"source": [ |
|
|
20 |
"%%time\n", |
|
|
21 |
"import pandas as pd\n", |
|
|
22 |
"import numpy as np\n", |
|
|
23 |
"from sklearn.ensemble import ExtraTreesClassifier\n", |
|
|
24 |
"from sklearn.metrics import classification_report\n", |
|
|
25 |
"from sklearn.model_selection import train_test_split\n", |
|
|
26 |
"from sklearn.ensemble import RandomForestClassifier\n", |
|
|
27 |
"from sklearn.linear_model import LogisticRegression\n", |
|
|
28 |
"from sklearn.tree import DecisionTreeClassifier\n", |
|
|
29 |
"from itertools import combinations \n", |
|
|
30 |
"from mlxtend.classifier import StackingClassifier\n", |
|
|
31 |
"from sklearn import model_selection\n", |
|
|
32 |
"\n" |
|
|
33 |
] |
|
|
34 |
}, |
|
|
35 |
{ |
|
|
36 |
"cell_type": "code", |
|
|
37 |
"execution_count": 4, |
|
|
38 |
"metadata": {}, |
|
|
39 |
"outputs": [ |
|
|
40 |
{ |
|
|
41 |
"name": "stdout", |
|
|
42 |
"output_type": "stream", |
|
|
43 |
"text": [ |
|
|
44 |
"CPU times: user 46.9 s, sys: 10.7 s, total: 57.6 s\n", |
|
|
45 |
"Wall time: 57.6 s\n" |
|
|
46 |
] |
|
|
47 |
} |
|
|
48 |
], |
|
|
49 |
"source": [ |
|
|
50 |
"%%time\n", |
|
|
51 |
"train = pd.read_csv(\"1_min_train.csv\")\n", |
|
|
52 |
"test = pd.read_csv(\"1_min_test.csv\")" |
|
|
53 |
] |
|
|
54 |
}, |
|
|
55 |
{ |
|
|
56 |
"cell_type": "code", |
|
|
57 |
"execution_count": 5, |
|
|
58 |
"metadata": {}, |
|
|
59 |
"outputs": [ |
|
|
60 |
{ |
|
|
61 |
"name": "stdout", |
|
|
62 |
"output_type": "stream", |
|
|
63 |
"text": [ |
|
|
64 |
"(2520000, 11)\n", |
|
|
65 |
"(28950603, 11)\n" |
|
|
66 |
] |
|
|
67 |
} |
|
|
68 |
], |
|
|
69 |
"source": [ |
|
|
70 |
"print(train.shape)\n", |
|
|
71 |
"print(test.shape)" |
|
|
72 |
] |
|
|
73 |
}, |
|
|
74 |
{ |
|
|
75 |
"cell_type": "code", |
|
|
76 |
"execution_count": 6, |
|
|
77 |
"metadata": {}, |
|
|
78 |
"outputs": [ |
|
|
79 |
{ |
|
|
80 |
"data": { |
|
|
81 |
"text/plain": [ |
|
|
82 |
"['chest_ACC_x',\n", |
|
|
83 |
" 'chest_ACC_y',\n", |
|
|
84 |
" 'chest_ACC_z',\n", |
|
|
85 |
" 'chest_ECG',\n", |
|
|
86 |
" 'chest_EMG',\n", |
|
|
87 |
" 'chest_EDA',\n", |
|
|
88 |
" 'chest_Temp',\n", |
|
|
89 |
" 'chest_Resp']" |
|
|
90 |
] |
|
|
91 |
}, |
|
|
92 |
"execution_count": 6, |
|
|
93 |
"metadata": {}, |
|
|
94 |
"output_type": "execute_result" |
|
|
95 |
} |
|
|
96 |
], |
|
|
97 |
"source": [ |
|
|
98 |
"features=train.columns.tolist()\n", |
|
|
99 |
"features = features[3:]\n", |
|
|
100 |
"features" |
|
|
101 |
] |
|
|
102 |
}, |
|
|
103 |
{ |
|
|
104 |
"cell_type": "code", |
|
|
105 |
"execution_count": 10, |
|
|
106 |
"metadata": {}, |
|
|
107 |
"outputs": [], |
|
|
108 |
"source": [ |
|
|
109 |
"clf1 = ExtraTreesClassifier(n_estimators=50, n_jobs=10, verbose=1,random_state=0)\n", |
|
|
110 |
"clf2 = DecisionTreeClassifier()\n", |
|
|
111 |
"clf3 = RandomForestClassifier(n_estimators=10)\n", |
|
|
112 |
"clf4 = LogisticRegression()" |
|
|
113 |
] |
|
|
114 |
}, |
|
|
115 |
{ |
|
|
116 |
"cell_type": "code", |
|
|
117 |
"execution_count": 14, |
|
|
118 |
"metadata": {}, |
|
|
119 |
"outputs": [], |
|
|
120 |
"source": [ |
|
|
121 |
"sclf = StackingClassifier(classifiers=[clf1, clf2, clf3], meta_classifier=clf4)" |
|
|
122 |
] |
|
|
123 |
}, |
|
|
124 |
{ |
|
|
125 |
"cell_type": "code", |
|
|
126 |
"execution_count": 18, |
|
|
127 |
"metadata": { |
|
|
128 |
"scrolled": true |
|
|
129 |
}, |
|
|
130 |
"outputs": [ |
|
|
131 |
{ |
|
|
132 |
"name": "stderr", |
|
|
133 |
"output_type": "stream", |
|
|
134 |
"text": [ |
|
|
135 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
136 |
"/home/sf/.local/lib/python3.6/site-packages/joblib/externals/loky/process_executor.py:706: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n", |
|
|
137 |
" \"timeout or by a memory leak.\", UserWarning\n", |
|
|
138 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 2.5min\n", |
|
|
139 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 4.2min finished\n", |
|
|
140 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
141 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 15.9s\n", |
|
|
142 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 23.4s finished\n", |
|
|
143 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
144 |
"/home/sf/.local/lib/python3.6/site-packages/joblib/externals/loky/process_executor.py:706: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n", |
|
|
145 |
" \"timeout or by a memory leak.\", UserWarning\n", |
|
|
146 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 1.3min\n", |
|
|
147 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 2.4min finished\n", |
|
|
148 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
149 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 15.5s\n", |
|
|
150 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 24.3s finished\n", |
|
|
151 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
152 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 1.4min\n", |
|
|
153 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 2.6min finished\n", |
|
|
154 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
155 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 16.7s\n", |
|
|
156 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 25.8s finished\n" |
|
|
157 |
] |
|
|
158 |
}, |
|
|
159 |
{ |
|
|
160 |
"name": "stdout", |
|
|
161 |
"output_type": "stream", |
|
|
162 |
"text": [ |
|
|
163 |
"Accuracy: 0.49 (+/- 0.10) [ExtraTreesClassifier]\n", |
|
|
164 |
"Accuracy: 0.38 (+/- 0.06) [DecisionTreeClassifier]\n", |
|
|
165 |
"Accuracy: 0.43 (+/- 0.10) [RandomForestClassifier]\n" |
|
|
166 |
] |
|
|
167 |
}, |
|
|
168 |
{ |
|
|
169 |
"name": "stderr", |
|
|
170 |
"output_type": "stream", |
|
|
171 |
"text": [ |
|
|
172 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
173 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 1.3min\n", |
|
|
174 |
"/home/sf/.local/lib/python3.6/site-packages/joblib/externals/loky/process_executor.py:706: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n", |
|
|
175 |
" \"timeout or by a memory leak.\", UserWarning\n", |
|
|
176 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 2.0min finished\n", |
|
|
177 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
178 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 20.1s\n", |
|
|
179 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 30.6s finished\n", |
|
|
180 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", |
|
|
181 |
" FutureWarning)\n", |
|
|
182 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n", |
|
|
183 |
" \"this warning.\", FutureWarning)\n", |
|
|
184 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
185 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 11.9s\n", |
|
|
186 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 17.0s finished\n", |
|
|
187 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
188 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 1.4min\n", |
|
|
189 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 2.7min finished\n", |
|
|
190 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
191 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 19.8s\n", |
|
|
192 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 37.8s finished\n", |
|
|
193 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", |
|
|
194 |
" FutureWarning)\n", |
|
|
195 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n", |
|
|
196 |
" \"this warning.\", FutureWarning)\n", |
|
|
197 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
198 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 14.4s\n", |
|
|
199 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 23.7s finished\n", |
|
|
200 |
"[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n", |
|
|
201 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 1.6min\n", |
|
|
202 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 2.8min finished\n", |
|
|
203 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
204 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 16.6s\n", |
|
|
205 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 26.6s finished\n", |
|
|
206 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", |
|
|
207 |
" FutureWarning)\n", |
|
|
208 |
"/home/sf/.local/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n", |
|
|
209 |
" \"this warning.\", FutureWarning)\n", |
|
|
210 |
"[Parallel(n_jobs=10)]: Using backend ThreadingBackend with 10 concurrent workers.\n", |
|
|
211 |
"[Parallel(n_jobs=10)]: Done 30 tasks | elapsed: 8.7s\n", |
|
|
212 |
"[Parallel(n_jobs=10)]: Done 50 out of 50 | elapsed: 13.6s finished\n" |
|
|
213 |
] |
|
|
214 |
}, |
|
|
215 |
{ |
|
|
216 |
"name": "stdout", |
|
|
217 |
"output_type": "stream", |
|
|
218 |
"text": [ |
|
|
219 |
"Accuracy: 0.37 (+/- 0.06) [LogisticRegression]\n" |
|
|
220 |
] |
|
|
221 |
} |
|
|
222 |
], |
|
|
223 |
"source": [ |
|
|
224 |
"for clf, label in zip([clf1, clf2, clf3, sclf], ['ExtraTreesClassifier','DecisionTreeClassifier','RandomForestClassifier','LogisticRegression']):\n", |
|
|
225 |
"\n", |
|
|
226 |
" scores = model_selection.cross_val_score(clf, test[features], test['target'],cv=3, scoring='accuracy')\n", |
|
|
227 |
" \n", |
|
|
228 |
" print(\"Accuracy: %0.2f (+/- %0.2f) [%s]\" % (scores.mean(), scores.std(), label))" |
|
|
229 |
] |
|
|
230 |
}, |
|
|
231 |
{ |
|
|
232 |
"cell_type": "code", |
|
|
233 |
"execution_count": null, |
|
|
234 |
"metadata": {}, |
|
|
235 |
"outputs": [], |
|
|
236 |
"source": [] |
|
|
237 |
} |
|
|
238 |
], |
|
|
239 |
"metadata": { |
|
|
240 |
"kernelspec": { |
|
|
241 |
"display_name": "Python 3", |
|
|
242 |
"language": "python", |
|
|
243 |
"name": "python3" |
|
|
244 |
}, |
|
|
245 |
"language_info": { |
|
|
246 |
"codemirror_mode": { |
|
|
247 |
"name": "ipython", |
|
|
248 |
"version": 3 |
|
|
249 |
}, |
|
|
250 |
"file_extension": ".py", |
|
|
251 |
"mimetype": "text/x-python", |
|
|
252 |
"name": "python", |
|
|
253 |
"nbconvert_exporter": "python", |
|
|
254 |
"pygments_lexer": "ipython3", |
|
|
255 |
"version": "3.5.2" |
|
|
256 |
} |
|
|
257 |
}, |
|
|
258 |
"nbformat": 4, |
|
|
259 |
"nbformat_minor": 2 |
|
|
260 |
} |