21/4/25, 8:57 p.m. Untitled1.

ipynb - Colab

import pandas as pd
import matplotlib.pyplot as plt
from sklearn.tree import DecisionTreeClassifier, plot_tree
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, confusion_matrix, ConfusionMatrixDisplay
from sklearn.decomposition import PCA

import ipywidgets as widgets

from IPython.display import display, clear_output

data = pd.read_csv('Pokemon.csv')
data = data[['HP', 'Attack', 'Defense', 'Sp. Atk', 'Sp. Def', 'Speed', 'Type 1']]
data = data.dropna()
data['Type 1'] = data['Type 1'].astype('category')
data['Type Code'] = data['Type 1'].cat.codes # Guardar códigos de tipo

X = data[['HP', 'Attack', 'Defense', 'Sp. Atk', 'Sp. Def', 'Speed']]

y = data['Type Code']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

type_names = data[['Type 1', 'Type Code']].drop_duplicates().sort_values('Type Code')['Type 1'].values

max_depth_slider = widgets.IntSlider(value=3, min=1, max=10, description='Max Depth:')

criterion_dropdown = widgets.Dropdown(
options=['gini', 'entropy'],
value='gini',
description='Criterion:'
)
button = widgets.Button(description="Generar Árbol", button_style='success')

def graficar_importancia(importances):
plt.figure()
plt.barh(X.columns, importances, color='skyblue')
plt.xlabel("Importancia")
plt.title("Importancia de Características")
plt.grid(True)
plt.show()

def graficar_precision_vs_profundidad(criterion):
depths = range(1, 11)
accuracies = []
for d in depths:
model = DecisionTreeClassifier(max_depth=d, criterion=criterion)
model.fit(X_train, y_train)
pred = model.predict(X_test)
acc = accuracy_score(y_test, pred)
accuracies.append(acc)

plt.figure()
plt.plot(depths, accuracies, marker='o', linestyle='--', color='orange')
plt.xlabel('Profundidad del Árbol')
plt.ylabel('Precisión')
plt.title('Precisión vs Profundidad del Árbol')
plt.grid(True)
plt.show()

def graficar_pca():
pca = PCA(n_components=2)
X_pca = pca.fit_transform(X)
plt.figure(figsize=(8,6))
scatter = plt.scatter(X_pca[:, 0], X_pca[:, 1], c=y, cmap='tab20', alpha=0.6)
lt l b l("C t P i i l 1")
https://colab.research.google.com/drive/1YuDHSG8QsmdJ8-w_ERtMrL9e-_U70TnP#scrollTo=CwfwYSiAKpUg&printMode=true 1/5
21/4/25, 8:57 p.m. Untitled1.ipynb - Colab
plt.xlabel("Componente Principal 1")
plt.ylabel("Componente Principal 2")
plt.title("Visualización PCA del dataset Pokémon")
plt.colorbar(scatter, label='Tipo')
plt.show()

def entrenar_y_graficar(max_depth, criterion):

clear_output(wait=True)
display(max_depth_slider, criterion_dropdown, button)

# Entrenar modelo
model = DecisionTreeClassifier(max_depth=max_depth, criterion=criterion)
model.fit(X_train, y_train)

# Evaluación
predictions = model.predict(X_test)
accuracy = accuracy_score(y_test, predictions)
print(f"\n✅ Precisión del modelo: {accuracy:.2f}")

# Importancia de características
importances = model.feature_importances_
graficar_importancia(importances)

# Matriz de Confusión (solo con clases presentes en test/predicción)

labels_presentes = sorted(set(y_test) | set(predictions))
nombres_presentes = [type_names[i] for i in labels_presentes]
cm = confusion_matrix(y_test, predictions, labels=labels_presentes)
disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=nombres_presentes)
disp.plot(cmap='Blues')
plt.title("Matriz de Confusión")
plt.xticks(rotation=90)
plt.show()

# Árbol de Decisión
if model.tree_.node_count > 1:
plt.figure(figsize=(20, 10))
plot_tree(model, feature_names=X.columns, class_names=type_names.astype(str), filled=True)
plt.title("Árbol de Decisión")
plt.show()
else:
print("⚠️ El árbol es demasiado simple para graficar. Aumenta la profundidad.")

# Gráfico adicional: Precisión vs Profundidad

graficar_precision_vs_profundidad(criterion)

# Gráfico adicional: PCA

graficar_pca()

def on_button_click(b):
plt.close('all') # Cierra gráficos anteriores
entrenar_y_graficar(max_depth_slider.value, criterion_dropdown.value)

button.on_click(on_button_click)

display(max_depth_slider, criterion_dropdown, button)

https://colab.research.google.com/drive/1YuDHSG8QsmdJ8-w_ERtMrL9e-_U70TnP#scrollTo=CwfwYSiAKpUg&printMode=true 2/5
21/4/25, 8:57 p.m. Untitled1.ipynb - Colab

Max Depth: 5

Criterion: gini

Generar Árbol

✅ Precisión del modelo: 0.18

https://colab.research.google.com/drive/1YuDHSG8QsmdJ8-w_ERtMrL9e-_U70TnP#scrollTo=CwfwYSiAKpUg&printMode=true 3/5
21/4/25, 8:57 p.m. Untitled1.ipynb - Colab

https://colab.research.google.com/drive/1YuDHSG8QsmdJ8-w_ERtMrL9e-_U70TnP#scrollTo=CwfwYSiAKpUg&printMode=true 4/5
21/4/25, 8:57 p.m. Untitled1.ipynb - Colab

https://colab.research.google.com/drive/1YuDHSG8QsmdJ8-w_ERtMrL9e-_U70TnP#scrollTo=CwfwYSiAKpUg&printMode=true 5/5