Curso de Fundamentos de IA: investigando algoritmos e abordagens de machine learning

Por Ricardo Costa Val do Rosario

Segui as instruções e encontrei o resultado abaixo. Após solicitei o CoPilot 365 para analisar o resultado encontrado Pergunto se está correto ?

Aula 02 Faça como eu fiz: ajustando modelos de machine learning

[1]

from sklearn.datasets import fetch_california_housing
from sklearn.model_selection import train_test_split

california = fetch_california_housing()
X = california.data
y = california.target


# Dividindo os dados entre treino e teste

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

[2]

# Passo 1: Importar as bibliotecas necessárias

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score


# Passo 2: Carregar o dataset Iris

iris = load_iris()
X = iris.data  # Características (comprimento e largura das pétalas e sépalas)
y = iris.target  # Rótulos (espécies das flores)


# Passo 3: Dividir os dados em treino e teste
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# Passo 4: Treinar o modelo

model = DecisionTreeClassifier()
model.fit(X_train, y_train)


# Passo 5: Fazer previsões e avaliar o modelo

y_pred = model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print(f"Acurácia do modelo: {accuracy * 100:.2f}%")
Acurácia do modelo: 100.00%

[3]

import pandas as pd
from sklearn.datasets import load_diabetes


# Carregar os dados

diabetes = load_diabetes()
X = diabetes.data
y = diabetes.target


# Dividir os dados entre treino e teste

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

[4]

from sklearn.ensemble import RandomForestRegressor
import matplotlib.pyplot as plt


# Treinar o modelo

model = RandomForestRegressor()
model.fit(X_train, y_train)


# Obter a importância das features

importances = model.feature_importances_
feature_names = diabetes.feature_names


# Criar um gráfico de barras

plt.figure(figsize=(10, 6))
plt.barh(feature_names, importances)
plt.title("Importância das Features")
plt.xlabel("Importância")
plt.ylabel("Feature")
plt.show()

[5]

# Passo 1: Importar as bibliotecas necessárias

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score


# Passo 2: Carregar o dataset Iris

iris = load_iris()
X = iris.data  # Características (comprimento e largura das pétalas e sépalas)
y = iris.target  # Rótulos (espécies das flores)

# Passo 3: Dividir os dados em treino e teste
X
_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# Passo 4: Treinar o modelo
model = DecisionTreeClassifier()
model.fit(X_train, y_train)

# Passo 5: Fazer previsões e avaliar o modelo
y_pred = model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print(f"Acurácia do modelo: {accuracy * 100:.2f}%")
Acurácia do modelo: 100.00%
[ ]
Type code here to execute.

Análise da IA CoPilot 365

• The selected code demonstrates a basic machine learning workflow using the Iris dataset and a Decision Tree Classifier.

Here's a breakdown of what each section does:

• Passo 1:

Importar as bibliotecas necessárias: 
This section imports the required libraries from the sklearn package, includingfunctions to load the Iris dataset, 
split data, use a Decision Tree classifier, and calculate accuracy.

• Passo 2:

Carregar o dataset Iris: 
This loads the Iris dataset, separating the features (measurements of the flowers) into X and the target labels (the species 
of the flowers) into y.

• Passo 3:

Dividir os dados em treino e teste: 
The data is split into training sets (X_train, y_train) and testing sets (X_test, y_test). 
The training data is used to train the model, and the testing data is used to evaluate its performance. test_size=0.3 
means 30% of the data is used for testing, and random_state=42 ensures the split is the same each time the code is run.

• Passo 4:

Treinar o modelo: 
A Decision Tree Classifier model is created and then trained using the training data.

• Passo 5:

Fazer previsões e avaliar o modelo: 
The trained model makes predictions on the test data (y_pred). 
The accuracy_score function then compares these predictions to the actual labels (y_test) to
calculate the accuracy of the model. 
Finally, the accuracy is printed to the console.