File size: 4,433 Bytes
e6b3df4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 |
# -*- coding: utf-8 -*-
"""Workshop-Linear kernel.ipynb
Automatically generated by Colab.
Original file is located at
https://colab.research.google.com/drive/1PlQo6At9kkwbacgmJdJ7FuN93WVLFoEY
"""
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
# Breast cancer dataset: target = malignant (0) / benign (1)
data = datasets.load_breast_cancer()
X = data.data # features
y = data.target # labels
print("Feature matrix shape:", X.shape)
print("Labels shape:", y.shape)
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size=0.3, random_state=42, stratify=y
)
print("Train size:", X_train.shape[0], "Test size:", X_test.shape[0])
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)
linear_svm = SVC(kernel='linear', C=1.0, random_state=42)
linear_svm.fit(X_train_scaled, y_train)
y_pred_linear = linear_svm.predict(X_test_scaled)
acc_linear = accuracy_score(y_test, y_pred_linear)
print("\n=== Linear Kernel SVM (Dot Product) ===")
print("Accuracy:", acc_linear)
print("Confusion matrix:\n", confusion_matrix(y_test, y_pred_linear))
print("Classification report:\n", classification_report(y_test, y_pred_linear))
rbf_svm = SVC(kernel='rbf', C=1.0, gamma='scale', random_state=42)
rbf_svm.fit(X_train_scaled, y_train)
y_pred_rbf = rbf_svm.predict(X_test_scaled)
acc_rbf = accuracy_score(y_test, y_pred_rbf)
print("\n=== RBF Kernel SVM ===")
print("Accuracy:", acc_rbf)
print("Confusion matrix:\n", confusion_matrix(y_test, y_pred_rbf))
print("Classification report:\n", classification_report(y_test, y_pred_rbf))
X_small = X_train_scaled[:5]
# Dot product kernel matrix: K_ij = x_i · x_j
K_dot = X_small @ X_small.T
# RBF kernel matrix by hand
def rbf_kernel(X1, X2, gamma):
# ||x - x'||^2 = (x^2 summed) + (x'^2 summed) - 2 x·x'
X1_sq = np.sum(X1 ** 2, axis=1).reshape(-1, 1)
X2_sq = np.sum(X2 ** 2, axis=1).reshape(1, -1)
sq_dist = X1_sq + X2_sq - 2 * X1 @ X2.T
return np.exp(-gamma * sq_dist)
gamma_value = 0.1
K_rbf = rbf_kernel(X_small, X_small, gamma=gamma_value)
print("\n=== Example Kernel Matrices on 5 Training Samples ===")
print("\nDot-product kernel matrix (linear):\n", K_dot)
print("\nRBF kernel matrix (gamma = 0.1):\n", K_rbf)
import gradio as gr
def predict_cancer(feature1, feature2, feature3, feature4, feature5, feature6, feature7, feature8, feature9, feature10, feature11, feature12, feature13, feature14, feature15, feature16, feature17, feature18, feature19, feature20, feature21, feature22, feature23, feature24, feature25, feature26, feature27, feature28, feature29, feature30):
# Collect all features into a numpy array
features = np.array([
feature1, feature2, feature3, feature4, feature5, feature6, feature7, feature8, feature9, feature10,
feature11, feature12, feature13, feature14, feature15, feature16, feature17, feature18, feature19, feature20,
feature21, feature22, feature23, feature24, feature25, feature26, feature27, feature28, feature29, feature30
]).reshape(1, -1)
# Scale the input features using the pre-trained scaler
features_scaled = scaler.transform(features)
# Get predictions from both models
prediction_linear = linear_svm.predict(features_scaled)[0]
prediction_rbf = rbf_svm.predict(features_scaled)[0]
# Map predictions to human-readable labels
label_map = {0: "Malignant", 1: "Benign"}
output_linear = label_map[prediction_linear]
output_rbf = label_map[prediction_rbf]
return output_linear, output_rbf
# Create Gradio input components for 30 features
inputs = [
gr.Number(label=f"Feature {i+1}") for i in range(30)
]
# Create Gradio output components for both models' predictions
outputs = [
gr.Textbox(label="Linear SVM Prediction"),
gr.Textbox(label="RBF SVM Prediction")
]
# Create and launch the Gradio interface
iface = gr.Interface(
fn=predict_cancer,
inputs=inputs,
outputs=outputs,
title="Breast Cancer Prediction using SVM",
description="Enter 30 features to predict if a tumor is Malignant (0) or Benign (1) using Linear and RBF Kernel SVMs."
)
iface.launch() |