Pra-2 Folder

Overview

The Pra-2 folder contains thirteen scripts and notebooks for advanced machine learning and data analysis tasks, covering classification, anomaly detection, clustering, and sequence modeling.

File: shape_classifier.py

Description: This script generates synthetic images of circles, squares, and triangles, adds noise, and uses a KNN classifier to classify them.

Dependencies: numpy, opencv-python, scikit-learn, matplotlib

Code Snippet:

                Copy
                import numpy as np
import cv2
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier

def generate_shape(shape, size=32):
    img = np.zeros((size, size), dtype=np.uint8)
    if shape == "circle":
        cv2.circle(img, (16,16), 10, 255, -1)
    elif shape == "square":
        cv2.rectangle(img, (8,8), (24,24), 255, -1)
    elif shape == "triangle":
        pts = np.array([[16,4],[4,28],[28,28]], np.int32)
        cv2.drawContours(img, [pts], 0, 255, -1)
    return img

def create_dataset(n=100):
    X, y, shapes = [], [], ["circle", "square", "triangle"]
    for idx, shape in enumerate(shapes):
        for _ in range(n):
            img = generate_shape(shape)
            noise = np.random.normal(0, 10, img.shape).astype(np.uint8)
            noisy_img = cv2.add(img, noise)
            X.append(noisy_img.flatten())
            y.append(idx)
    return np.array(X), np.array(y), shapes
            

Other Files

• 1.ipynb, 1.py: Decision tree classifier for Iris dataset.
• anomaly_detection.py: Isolation Forest for anomaly detection on KDD Cup dataset.
• dna_seq.py: HMM for DNA sequence state prediction.
• gmm_digits.py: Gaussian Mixture Model for digit clustering.
• iris_classification.py: Random forest classifier for Iris dataset with PCA visualization.
• knn_iris.py: KNN classifier for Iris dataset with k-value optimization.
• mnist_clssification.py: Gaussian Naive Bayes for MNIST digit classification.
• test.py: Empty file, possibly a placeholder.
• textClassificationModel.py: Text classification pipeline for 20 Newsgroups dataset.
• weather_hmm.py: HMM for weather state prediction.
• wine_quality_analysis.py: Statistical analysis of synthetic wine data.