📑 Sklearn Workout 01. Partial Dependence & Regression

sklearn_01.html

✒️   Modules & Settings

xxxxxxxxxx

 

1

import sklearn as sl,pandas as pd,numpy as np,pylab as pl

2

from sklearn.inspection import plot_partial_dependence

3

from sklearn.experimental import enable_hist_gradient_boosting

4

from sklearn.ensemble import HistGradientBoostingRegressor

Language: Sage Gap GP HTML Macaulay2 Maxima Octave Python R Singular

---

Messages

✒️   Partial Dependence & Regression

xxxxxxxxxx

 

1

n_samples=1000; rng=np.random.RandomState(123)

2

X=rng.randn(n_samples,3)

3

noise=rng.normal(loc=0.,scale=.5,size=n_samples)

4

y=(2*X[:,0]+np.sin(3*np.pi*X[:,0])-noise)

5

gbdt_no_cst=HistGradientBoostingRegressor().fit(X,y)

6

gbdt_cst=HistGradientBoostingRegressor(

7

    monotonic_cst=[1,0,0]).fit(X,y)

8

pl.rcParams['figure.figsize']=(6,3)

9

disp=plot_partial_dependence(

10

    gbdt_no_cst,X,features=[0],

11

    feature_names=['feature 0'],

12

    line_kw={'linewidth':2,

13

             'color':'darkred',

14

             'label':'unconstrained'})

15

plot_partial_dependence(

16

    gbdt_cst,X,features=[0],ax=disp.axes_,

17

    line_kw={'linewidth':2,

18

             'color':'darkblue',

19

             'label':'constrained'})

20

disp.axes_[0,0].plot(

21

    X[:,0],y,'o',ms=1,c='darkorchid',

22

    alpha=.7,zorder=-1,label='samples')

23

disp.axes_[0,0].set_ylim(-5,5)

24

disp.axes_[0,0].set_xlim(-1,1)

25

pl.legend(); pl.grid(); pl.show()

Language: Sage Gap GP HTML Macaulay2 Maxima Octave Python R Singular

---

Messages