{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "Написать модель Random Forest на датасет больных сердечными болезнями\n", "Определить, какой атрибут оказывает наибольшее влияние на наличие или отсутствие болезни. \n", "Выдвинуть и оценить гипотезу на самый важный атрибут, \n", "полученный из feature_importance (Random_Forest). \n", "Сделать выводы" ] }, { "cell_type": "code", "execution_count": 101, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import pandas_profiling \n", "import numpy as np \n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "\n", "from sklearn import preprocessing\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import classification_report\n", "from sklearn.model_selection import train_test_split\n", "\n", "import scipy.stats as sts\n", "import statistics\n", "\n", "import researchpy as rp\n", "\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 102, "metadata": {}, "outputs": [ { "data": { "text/html": [ "

\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "

	id	age	gender	height	weight	ap_hi	ap_lo	cholesterol	gluc	smoke	alco	active	cardio
count	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.00	70000.0	70000.0
mean	49972.42	19468.87	1.35	164.36	74.21	128.82	96.63	1.37	1.23	0.09	0.05	0.8	0.5
std	28851.30	2467.25	0.48	8.21	14.40	154.01	188.47	0.68	0.57	0.28	0.23	0.4	0.5
min	0.00	10798.00	1.00	55.00	10.00	-150.00	-70.00	1.00	1.00	0.00	0.00	0.0	0.0
25%	25006.75	17664.00	1.00	159.00	65.00	120.00	80.00	1.00	1.00	0.00	0.00	1.0	0.0
50%	50001.50	19703.00	1.00	165.00	72.00	120.00	80.00	1.00	1.00	0.00	0.00	1.0	0.0
75%	74889.25	21327.00	2.00	170.00	82.00	140.00	90.00	2.00	1.00	0.00	0.00	1.0	1.0
max	99999.00	23713.00	2.00	250.00	200.00	16020.00	11000.00	3.00	3.00	1.00	1.00	1.0	1.0

\n", "

" ], "text/plain": [ " id age gender height weight ap_hi ap_lo \\\n", "count 70000.00 70000.00 70000.00 70000.00 70000.00 70000.00 70000.00 \n", "mean 49972.42 19468.87 1.35 164.36 74.21 128.82 96.63 \n", "std 28851.30 2467.25 0.48 8.21 14.40 154.01 188.47 \n", "min 0.00 10798.00 1.00 55.00 10.00 -150.00 -70.00 \n", "25% 25006.75 17664.00 1.00 159.00 65.00 120.00 80.00 \n", "50% 50001.50 19703.00 1.00 165.00 72.00 120.00 80.00 \n", "75% 74889.25 21327.00 2.00 170.00 82.00 140.00 90.00 \n", "max 99999.00 23713.00 2.00 250.00 200.00 16020.00 11000.00 \n", "\n", " cholesterol gluc smoke alco active cardio \n", "count 70000.00 70000.00 70000.00 70000.00 70000.0 70000.0 \n", "mean 1.37 1.23 0.09 0.05 0.8 0.5 \n", "std 0.68 0.57 0.28 0.23 0.4 0.5 \n", "min 1.00 1.00 0.00 0.00 0.0 0.0 \n", "25% 1.00 1.00 0.00 0.00 1.0 0.0 \n", "50% 1.00 1.00 0.00 0.00 1.0 0.0 \n", "75% 2.00 1.00 0.00 0.00 1.0 1.0 \n", "max 3.00 3.00 1.00 1.00 1.0 1.0 " ] }, "execution_count": 102, "metadata": {}, "output_type": "execute_result" } ], "source": [ "url = \"http://yustiks.ru/dataset/cardio_train.csv\"\n", "data = pd.read_csv(url,sep=\";\")\n", "data.isnull().any()\n", "data.describe().round(2)" ] }, { "cell_type": "code", "execution_count": 103, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "

Number of variables	13
Number of observations	68985
Total Missing (%)	0.0%
Total size in memory	6.8 MiB
Average record size in memory	104.0 B

Numeric	8
Categorical	0
Boolean	5
Date	0
Text (Unique)	0
Rejected	0
Unsupported	0

Distinct count	68985
Unique (%)	100.0%
Missing (%)	0.0%
Missing (n)	0
Infinite (%)	0.0%
Infinite (n)	0

Mean	34996
Minimum	0
Maximum	69999
Zeros (%)	0.0%

Minimum	0
5-th percentile	3503.2
Q1	17493
Median	35003
Q3	52483
95-th percentile	66497
Maximum	69999
Range	69999
Interquartile range	34990

Overview

Variables

Correlations

Sample

	age	gender	height	weight	ap_hi	ap_lo	cholesterol	gluc	smoke	alco	active	cardio
69995	52.676249	2	168	76.0	120	80	1	1	1	0	1	0
69996	61.878166	1	158	126.0	140	90	2	2	0	0	1	1
69997	52.199863	2	183	105.0	180	90	3	1	0	1	0	1
69998	61.412731	1	163	72.0	135	80	1	2	0	0	0	1
69999	56.235455	1	170	72.0	120	80	2	1	0	0	1	0

Standard deviation	20205
Coef of variation	0.57735
Kurtosis	-1.1997
Mean	34996
MAD	17497
Skewness	9.3765e-06
Sum	2414203412
Variance	408240000
Memory size	539.1 KiB

Value	Count	Frequency (%)
2047	1	0.0%	\n", " \n", "
2708	1	0.0%	\n", " \n", "
43712	1	0.0%	\n", " \n", "
66229	1	0.0%	\n", " \n", "
68276	1	0.0%	\n", " \n", "
21151	1	0.0%	\n", " \n", "
23198	1	0.0%	\n", " \n", "
17053	1	0.0%	\n", " \n", "
19100	1	0.0%	\n", " \n", "
29339	1	0.0%	\n", " \n", "
Other values (68975)	68975	100.0%	\n", " \n", "

Distinct count	8066
Unique (%)	11.7%
Missing (%)	0.0%
Missing (n)	0
Infinite (%)	0.0%
Infinite (n)	0

Minimum	29.563
5-th percentile	41.227
Q1	48.339
Median	53.938
Q3	58.379
95-th percentile	63.671
Maximum	64.923
Range	35.359
Interquartile range	10.04

Standard deviation	6.7574
Coef of variation	0.12681
Kurtosis	-0.82564
Mean	53.289
MAD	5.6341
Skewness	-0.30561
Sum	3676100
Variance	45.663
Memory size	539.1 KiB

1	\n", " \n", " 44932\n", " \n", " \n", "
2	\n", " \n", " 24053\n", " \n", " \n", "

Distinct count	108
Unique (%)	0.2%
Missing (%)	0.0%
Missing (n)	0
Infinite (%)	0.0%
Infinite (n)	0

Standard deviation	8.2043
Coef of variation	0.049917
Kurtosis	8.0763
Mean	164.36
MAD	6.3158
Skewness	-0.65477
Sum	11338352
Variance	67.31
Memory size	539.1 KiB

Distinct count	281
Unique (%)	0.4%
Missing (%)	0.0%
Missing (n)	0
Infinite (%)	0.0%
Infinite (n)	0

Value	Count	Frequency (%)
65.0	3814	5.5%	\n", " \n", "
70.0	3721	5.4%	\n", " \n", "
68.0	2799	4.1%	\n", " \n", "
75.0	2704	3.9%	\n", " \n", "
60.0	2688	3.9%	\n", " \n", "
80.0	2589	3.8%	\n", " \n", "
72.0	2275	3.3%	\n", " \n", "
69.0	2171	3.1%	\n", " \n", "
78.0	2050	3.0%	\n", " \n", "
74.0	1841	2.7%	\n", " \n", "
Other values (271)	42333	61.4%	\n", " \n", "

Standard deviation	14.329
Coef of variation	0.19332
Kurtosis	2.5578
Mean	74.119
MAD	10.995
Skewness	1.0078
Sum	5113100
Variance	205.32
Memory size	539.1 KiB

Standard deviation	17.699
Coef of variation	0.1401
Kurtosis	5.262
Mean	126.33
MAD	13.178
Skewness	0.14143
Sum	8714532
Variance	313.24
Memory size	539.1 KiB

Distinct count	93
Unique (%)	0.1%
Missing (%)	0.0%
Missing (n)	0
Infinite (%)	0.0%
Infinite (n)	0

Standard deviation	9.8057
Coef of variation	0.12054
Kurtosis	5.1305
Mean	81.35
MAD	6.7902
Skewness	0.49604
Sum	5611963
Variance	96.151
Memory size	539.1 KiB

Standard deviation	0.67869
Coef of variation	0.49743
Kurtosis	1.0291
Mean	1.3644
MAD	0.54666
Skewness	1.5979
Sum	94122
Variance	0.46062
Memory size	539.1 KiB

Standard deviation	0.57182
Coef of variation	0.46646
Kurtosis	4.3168
Mean	1.2259
MAD	0.38407
Skewness	2.4023
Sum	84567
Variance	0.32698
Memory size	539.1 KiB

0	\n", " \n", " 62924\n", " \n", " \n", "
1	\n", " \n", " \n", " \n", " 6061\n", "

0	\n", " \n", " 65288\n", " \n", " \n", "
1	\n", " \n", " \n", " \n", " 3697\n", "

1	\n", " \n", " 55414\n", " \n", " \n", "
0	\n", " \n", " 13571\n", " \n", " \n", "

0	\n", " \n", " 34844\n", " \n", " \n", "
1	\n", " \n", " 34141\n", " \n", " \n", "

	age	gender	height	weight	ap_hi	ap_lo	cholesterol	gluc	active	cardio
0	50.357290	2	168	62.0	110	80	1	1	1	0
1	55.381246	1	156	85.0	140	90	3	1	1	1
2	51.627652	1	165	64.0	130	70	3	1	0	1
3	48.249144	2	169	82.0	150	100	1	1	1	1
4	47.841205	1	156	56.0	100	60	1	1	0	0