using sklearn.datasets.fetch_openml¶
https://scikit-learn.org/stable/modules/generated/sklearn.datasets.fetch_openml.html
In [70]:
from sklearn.datasets import fetch_openml
import pandas as pd
import numpy as np
import statsmodels.formula.api as smf
import statsmodels.api as sm
In [40]:
X, y = fetch_openml("titanic", version=1, as_frame=True, return_X_y=True)
type(X)
Out[40]:
pandas.core.frame.DataFrame
In [41]:
X.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 1309 entries, 0 to 1308 Data columns (total 13 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 pclass 1309 non-null float64 1 name 1309 non-null object 2 sex 1309 non-null category 3 age 1046 non-null float64 4 sibsp 1309 non-null float64 5 parch 1309 non-null float64 6 ticket 1309 non-null object 7 fare 1308 non-null float64 8 cabin 295 non-null object 9 embarked 1307 non-null category 10 boat 486 non-null object 11 body 121 non-null float64 12 home.dest 745 non-null object dtypes: category(2), float64(6), object(5) memory usage: 115.4+ KB
In [42]:
y.dtype
Out[42]:
CategoricalDtype(categories=['0', '1'], ordered=False)
In [43]:
y.attrs
Out[43]:
{}
In [44]:
y.values
Out[44]:
['1', '1', '0', '0', '0', ..., '0', '0', '0', '0', '0'] Length: 1309 Categories (2, object): ['0', '1']
In [45]:
y
Out[45]:
0 1
1 1
2 0
3 0
4 0
..
1304 0
1305 0
1306 0
1307 0
1308 0
Name: survived, Length: 1309, dtype: category
Categories (2, object): ['0', '1']
In [46]:
y.index
Out[46]:
RangeIndex(start=0, stop=1309, step=1)
In [47]:
y.empty
Out[47]:
False
In [48]:
X.columns
Out[48]:
Index(['pclass', 'name', 'sex', 'age', 'sibsp', 'parch', 'ticket', 'fare',
'cabin', 'embarked', 'boat', 'body', 'home.dest'],
dtype='object')
In [49]:
y.name
Out[49]:
'survived'
In [50]:
y.flags
Out[50]:
<Flags(allows_duplicate_labels=True)>
In [51]:
y.astype
Out[51]:
<bound method NDFrame.astype of 0 1
1 1
2 0
3 0
4 0
..
1304 0
1305 0
1306 0
1307 0
1308 0
Name: survived, Length: 1309, dtype: category
Categories (2, object): ['0', '1']>
In [52]:
y.rank
Out[52]:
<bound method NDFrame.rank of 0 1
1 1
2 0
3 0
4 0
..
1304 0
1305 0
1306 0
1307 0
1308 0
Name: survived, Length: 1309, dtype: category
Categories (2, object): ['0', '1']>
In [53]:
y.type
--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) <ipython-input-53-2727d4ac428e> in <module>() ----> 1 y.type /usr/local/lib/python3.7/dist-packages/pandas/core/generic.py in __getattr__(self, name) 5485 ): 5486 return self[name] -> 5487 return object.__getattribute__(self, name) 5488 5489 def __setattr__(self, name: str, value) -> None: AttributeError: 'Series' object has no attribute 'type'
In [54]:
y.astype
Out[54]:
<bound method NDFrame.astype of 0 1
1 1
2 0
3 0
4 0
..
1304 0
1305 0
1306 0
1307 0
1308 0
Name: survived, Length: 1309, dtype: category
Categories (2, object): ['0', '1']>
In [55]:
type(y)
Out[55]:
pandas.core.series.Series
In [56]:
df = pd.merge(X, y, right_index = True, left_index = True)
print(df)
pclass name sex \
0 1.0 Allen, Miss. Elisabeth Walton female
1 1.0 Allison, Master. Hudson Trevor male
2 1.0 Allison, Miss. Helen Loraine female
3 1.0 Allison, Mr. Hudson Joshua Creighton male
4 1.0 Allison, Mrs. Hudson J C (Bessie Waldo Daniels) female
... ... ... ...
1304 3.0 Zabour, Miss. Hileni female
1305 3.0 Zabour, Miss. Thamine female
1306 3.0 Zakarian, Mr. Mapriededer male
1307 3.0 Zakarian, Mr. Ortin male
1308 3.0 Zimmerman, Mr. Leo male
age sibsp parch ticket fare cabin embarked boat body \
0 29.0000 0.0 0.0 24160 211.3375 B5 S 2 NaN
1 0.9167 1.0 2.0 113781 151.5500 C22 C26 S 11 NaN
2 2.0000 1.0 2.0 113781 151.5500 C22 C26 S None NaN
3 30.0000 1.0 2.0 113781 151.5500 C22 C26 S None 135.0
4 25.0000 1.0 2.0 113781 151.5500 C22 C26 S None NaN
... ... ... ... ... ... ... ... ... ...
1304 14.5000 1.0 0.0 2665 14.4542 None C None 328.0
1305 NaN 1.0 0.0 2665 14.4542 None C None NaN
1306 26.5000 0.0 0.0 2656 7.2250 None C None 304.0
1307 27.0000 0.0 0.0 2670 7.2250 None C None NaN
1308 29.0000 0.0 0.0 315082 7.8750 None S None NaN
home.dest survived
0 St Louis, MO 1
1 Montreal, PQ / Chesterville, ON 1
2 Montreal, PQ / Chesterville, ON 0
3 Montreal, PQ / Chesterville, ON 0
4 Montreal, PQ / Chesterville, ON 0
... ... ...
1304 None 0
1305 None 0
1306 None 0
1307 None 0
1308 None 0
[1309 rows x 14 columns]
In [57]:
df.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 1309 entries, 0 to 1308 Data columns (total 14 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 pclass 1309 non-null float64 1 name 1309 non-null object 2 sex 1309 non-null category 3 age 1046 non-null float64 4 sibsp 1309 non-null float64 5 parch 1309 non-null float64 6 ticket 1309 non-null object 7 fare 1308 non-null float64 8 cabin 295 non-null object 9 embarked 1307 non-null category 10 boat 486 non-null object 11 body 121 non-null float64 12 home.dest 745 non-null object 13 survived 1309 non-null category dtypes: category(3), float64(6), object(5) memory usage: 116.8+ KB
In [58]:
df.index
Out[58]:
RangeIndex(start=0, stop=1309, step=1)
$$ logit(Survival) = \alpha + \beta_{1}*PClass + \beta_3*Age + \epsilon $$
Here we are going to call a formula and use the GLM functionality
In [63]:
formula = 'survived ~ age + pclass'
model = smf.glm(formula = formula, data=df, family=sm.families.Binomial())
result=model.fit()
print(result.summary())
Generalized Linear Model Regression Results
==========================================================================================
Dep. Variable: ['survived[0]', 'survived[1]'] No. Observations: 1046
Model: GLM Df Residuals: 1043
Model Family: Binomial Df Model: 2
Link Function: logit Scale: 1.0000
Method: IRLS Log-Likelihood: -628.09
Date: Sun, 29 May 2022 Deviance: 1256.2
Time: 02:20:23 Pearson chi2: 1.05e+03
No. Iterations: 4
Covariance Type: nonrobust
==============================================================================
coef std err z P>|z| [0.025 0.975]
------------------------------------------------------------------------------
Intercept -3.1143 0.327 -9.520 0.000 -3.755 -2.473
age 0.0370 0.005 6.770 0.000 0.026 0.048
pclass 1.1015 0.096 11.530 0.000 0.914 1.289
==============================================================================
Interpretation¶
- Age - for each year increase in age, the odds of survival on the titanic decreased about 1-0.96 = 0.04 or 4%
- Pclass - for each marginal decrease in passenger class (ie 1st to 2nd class) the odds of survival decreased by 1-0.29 = 71%