Week 8 Friday

import pandas as pd
import altair as alt
from matplotlib import pyplot as plt

df = pd.DataFrame([
    [1,9,0],
    [8,9,2],
    [7,6,7],
    [11,6,4],
    [9,-3,1]
], columns=["x1", "x2", "y"])

base = alt.Chart(df).encode(
    x="x1",
    y="x2",
)

points = base.mark_circle()
labels = base.mark_text(dx=10).encode(
    text="y"
)

points+labels

from sklearn.tree import DecisionTreeRegressor

tree = DecisionTreeRegressor()

type(tree)

sklearn.tree._classes.DecisionTreeRegressor

tree.fit(df[["x1", "x2"]], df["y"])

DecisionTreeRegressor()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

from sklearn.tree import plot_tree

fig = plt.figure(figsize=(8,3))
_ = plot_tree(tree, 
                   feature_names=tree.feature_names_in_,
                   filled=False)

import numpy as np

rng = np.random.default_rng()

arr = rng.random(size=(5000,2))
arr[:, 0] *= 12
arr[:, 1] = 15*arr[:, 1] - 5

df_grid = pd.DataFrame(arr, columns=["x1", "x2"])
df_grid["pred"] = tree.predict(df_grid[["x1", "x2"]])

alt.Chart(df_grid).mark_circle().encode(
    x="x1",
    y="x2",
    color="pred:N"
)

• Two “manual” ways to control overfitting of decision trees in scikit-learn.

fig = plt.figure(figsize=(8,3))
_ = plot_tree(tree, 
                   feature_names=tree.feature_names_in_,
                   filled=False)

Created in Deepnote