carlleston

import sklearn.datasets as datasets
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import Pipeline
from sklearn.ensemble import RandomForestRegressor
from sklearn.decomposition import PCA

carlleston

df = pd.read_csv('df.csv', sep = ',')
df.head()

carlleston

X = df[['GDP','UnemploymentRate']]
y = df['Revenue']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,  random_state=0)

carlleston

regr = RandomForestRegressor(random_state = 100,bootstrap = True, max_depth=2,max_features=2,min_samples_leaf=3,min_samples_split=5,n_estimators=3)
pipe = Pipeline([
        ('scaler', StandardScaler()),
        ('reduce_dim', PCA()),
        ('regressor', regr)
        ])
pipe.fit(X_train,y_train)
ypipe=pipe.predict(X_test)

carlleston

conda install pydotplus
conda install graphviz

carlleston

from sklearn.externals.six import StringIO  
from IPython.display import Image  
from sklearn.tree import export_graphviz
import pydotplus
import os

os.environ['PATH'] = os.environ['PATH']+';'+os.environ['CONDA_PREFIX']+r"\Library\bin\graphviz"
dot_data = StringIO()
export_graphviz(pipe.named_steps['regressor'].estimators_[0], out_file=dot_data)
graph = pydotplus.graph_from_dot_data(dot_data.getvalue()) 

carlleston

carlleston

carlleston

carlleston