Scribd
Upload
39 views3 pages

PCA and Model Evaluation in Python

The document outlines a data analysis workflow using Python libraries to load datasets, visualize fault distributions, and preprocess data for machine learning. It employs PCA for dimensionality reduction and trains Random Forest and SVM models, evaluating their performance with confusion matrices and classification reports. The results highlight the accuracy and effectiveness of the models in classifying faults in the dataset.

Uploaded by

20112042paren
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
39 views3 pages

PCA and Model Evaluation in Python

The document outlines a data analysis workflow using Python libraries to load datasets, visualize fault distributions, and preprocess data for machine learning. It employs PCA for dimensionality reduction and trains Random Forest and SVM models, evaluating their performance with confusion matrices and classification reports. The results highlight the accuracy and effectiveness of the models in classifying faults in the dataset.

Uploaded by

20112042paren
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd

import pandas as pd

import numpy as np
import [Link] as plt
import seaborn as sns
from [Link] import PCA
from sklearn.model_selection import train_test_split, cross_val_score, ShuffleSplit
from [Link] import StandardScaler, LabelEncoder
from [Link] import RandomForestClassifier
from [Link] import SVC
from [Link] import accuracy_score, classification_report, confusion_matrix
from mpl_toolkits.mplot3d import Axes3D

# Set plot styles


[Link]({'[Link]': 14, "[Link]": "Times New Roman"})

# ----------------------
# Load Datasets
# ----------------------
df1 = pd.read_csv(r'/[Link], /[Link], /[Link]')
df2 = pd.read_csv(r'/[Link]')

# ----------------------
# Data Visualization: Fault Distribution
# ----------------------
[Link](figsize=(5, 5))
[Link](df1['label'].value_counts(), labels=df1['label'].unique(), autopct='%2.1f%
%', colors=sns.color_palette("pastel"))
[Link]("Fault Distribution in Limited Power Dataset")
[Link]()

# ----------------------
# Feature Correlation Heatmap
# ----------------------
X = [Link][:, 1:-1] # Features
Y = [Link][:, -1] # Target Labels

[Link](figsize=(12, 10))
[Link]([Link](), annot=True, cmap='coolwarm', fmt=".2f")
[Link]("Feature Correlation Heatmap")
[Link]()

# ----------------------
# Data Preprocessing
# ----------------------
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.2,
shuffle=True, random_state=42)
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = [Link](X_test)

# ----------------------
# PCA: Dimensionality Reduction
# ----------------------
for i in range(1, 14):
pca = PCA(n_components=i)
[Link](X_train_scaled)
print(f"Cumulative explained variance for {i} components:
{[Link](pca.explained_variance_ratio_):.4f}")
# Apply PCA with 2 components
pca_2 = PCA(n_components=2)
X_pca_2D = pca_2.fit_transform(X_train_scaled)
principalDf = [Link](data=X_pca_2D, columns=['PC1', 'PC2'])
principalDf['Fault'] = [Link](y_train)

[Link](figsize=(10, 8))
[Link](x=principalDf['PC1'], y=principalDf['PC2'],
hue=principalDf['Fault'], palette='Dark2', alpha=0.7)
[Link]("2D PCA Visualization")
[Link]("Principal Component 1")
[Link]("Principal Component 2")
[Link]()

# 3D PCA Visualization
pca_3 = PCA(n_components=3)
X_pca_3D = pca_3.fit_transform(X_train_scaled)
principalDf = [Link](data=X_pca_3D, columns=['PC1', 'PC2', 'PC3'])
principalDf['Fault'] = [Link](y_train)

fig = [Link](figsize=(10, 10))


ax = fig.add_subplot(111, projection='3d')
colors = sns.color_palette("husl", len(y_train.unique()))
for fault, color in zip(y_train.unique(), colors):
indices = principalDf['Fault'] == fault
[Link]([Link][indices, 'PC1'],
[Link][indices, 'PC2'],
[Link][indices, 'PC3'],
c=[color], label=fault, s=50)
ax.set_xlabel('Principal Component 1')
ax.set_ylabel('Principal Component 2')
ax.set_zlabel('Principal Component 3')
ax.set_title('3D PCA Visualization')
[Link]()
[Link]()

# ----------------------
# Train & Evaluate Models
# ----------------------
rf_model = RandomForestClassifier(n_estimators=100, random_state=42)
rf_model.fit(X_train_scaled, y_train)
y_pred_rf = rf_model.predict(X_test_scaled)

svm_model = SVC(kernel='rbf', C=1.0, gamma='scale', random_state=42)


svm_model.fit(X_train_scaled, y_train)
y_pred_svm = svm_model.predict(X_test_scaled)

# ----------------------
# Confusion Matrix Plot
# ----------------------
def plot_confusion_matrix(y_true, y_pred, model_name):
cm = confusion_matrix(y_true, y_pred)
[Link](figsize=(8, 6))
[Link](cm, annot=True, fmt="d", cmap="viridis",
xticklabels=[Link](y_true), yticklabels=[Link](y_true))
[Link]("Predicted Label")
[Link]("True Label")
[Link](f"Confusion Matrix - {model_name}")
[Link]()
plot_confusion_matrix(y_test, y_pred_rf, "Random Forest")
plot_confusion_matrix(y_test, y_pred_svm, "SVM")

# ----------------------
# Final Model Evaluations
# ----------------------
print("\nRandom Forest Results:")
print(f"Accuracy: {accuracy_score(y_test, y_pred_rf):.4f}")
print(classification_report(y_test, y_pred_rf))

print("\nSVM Results:")
print(f"Accuracy: {accuracy_score(y_test, y_pred_svm):.4f}")
print(classification_report(y_test, y_pred_svm))

You might also like