One Tree or a Crowd? — medium Machine Learning in Code problem | Incognition

One Tree or a Crowd?

10 pts · 10 coins

Machine Learning in CodeMedium

One Tree or a Crowd?

A single fully grown tree memorises. A forest of them votes, and their different mistakes cancel out. You want to see how many trees it takes.

Write tree_versus_forests(sizes) for the wine data.

Load the data with load_wine(return_X_y=True, as_frame=True) and split it with train_test_split(X, y, test_size=0.25, random_state=0, stratify=y).
Fit one DecisionTreeClassifier(random_state=0) and score it on the test rows.
For each value in sizes, in order, fit RandomForestClassifier(n_estimators=value, random_state=0) and score it on the test rows.

Return a dict:

"tree": the single tree's test score, rounded to 3 places.
"forests": a list of the forests' test scores, in the order of sizes, each rounded to 3 places.

Examples

sizes = [5, 25]

returns {'tree': 0.889, 'forests': [0.978, 1.0]}

sizes = [100]

returns {'tree': 0.889, 'forests': [1.0]}

Hints

solution.py

1from sklearn.datasets import load_wine
2from sklearn.ensemble import RandomForestClassifier
3from sklearn.model_selection import train_test_split
4from sklearn.tree import DecisionTreeClassifier
5
6
7def tree_versus_forests(sizes):
8    pass
9