miniproject-2
August 6, 2024
[13]: import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix, precision_score,␣
↪recall_score,f1_score
# Load the breast_cancer_dataset
df = pd.read_csv("C:\\Users\\SRIRAM\\OneDrive\\Documents\\WIN-SEM␣
↪2022-23\\B2_ML\\LAB\\PROJECT\\Breast_cancer.csv")
# column names of the dataset
column_names = ['Sample_code_number', 'Clump_Thickness', 'Uniformity_of_Cell␣
↪Size', 'Uniformity_of_Cell_Shape',
'Marginal_Adhesion', 'Single_Epithelial_Cell_Size',␣
↪'Bare_Nuclei', 'Bland_Chromatin',
'Normal_Nucleoli', 'Mitoses', 'Class']
df.columns = column_names
#changing values of target function from [2,4] to [0,1]
df['Class'] = np.where(df['Class'] == 2, 0, 1)
# Droping 'Sample code number' column which is not required for classification
df.drop('Sample_code_number', axis=1, inplace=True)
# Replace missing values (denoted by '?') with NaN
df.replace('?', np.nan, inplace=True)
# Droping the rows with missing values
df.dropna(inplace=True)
# Converting columns to numeric data type
df = df.astype({'Bare_Nuclei': 'int64', 'Class': 'int64'})
#updated df is shown as output
1
print("Updated data set without sample_code_number \n",df)
# Split the dataset into features and target
X = df.iloc[:, :-1] # loading Features into x
y = df.iloc[:, -1] # Target into y
# Split the dataset into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,␣
↪random_state=42)
# Standardize the features
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)
Updated data set without sample_code_number
Clump_Thickness Uniformity_of_Cell Size Uniformity_of_Cell_Shape \
0 5 1 1
1 5 4 4
2 3 1 1
3 6 8 8
4 4 1 1
.. … … …
694 3 1 1
695 2 1 1
696 5 10 10
697 4 8 6
698 4 8 8
Marginal_Adhesion Single_Epithelial_Cell_Size Bare_Nuclei \
0 1 2 1
1 5 7 10
2 1 2 2
3 1 3 4
4 3 2 1
.. … … …
694 1 3 2
695 1 2 1
696 3 7 3
697 4 3 4
698 5 4 5
Bland_Chromatin Normal_Nucleoli Mitoses Class
0 3 1 1 0
1 3 2 1 0
2 3 1 1 0
3 3 7 1 0
2
4 3 1 1 0
.. … … … …
694 1 1 1 0
695 1 1 1 0
696 8 10 2 1
697 10 6 1 1
698 10 4 1 1
[683 rows x 10 columns]
[14]: # Naive Bayes
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import accuracy_score
# Train the model
nb = GaussianNB()
nb.fit(X_train, y_train)
# Make predictions
y_pred = nb.predict(X_test)
# Compute confusion matrix for Naive bayes
sns.heatmap(confusion_matrix(y_test, y_pred), annot=True, cmap='Blues')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion matrix for testing set using Naive bayes')
plt.show()
print("\n")
# Evaluate performance(Accuracy)
nb_accuracy = accuracy_score(y_test, y_pred)
print("Naïve Bayesian Classifier accuracy:", nb_accuracy)
print("\n")
# Compute the precision
nb_precision = precision_score(y_test, y_pred)
print("Naive Bayes precision:", nb_precision)
print("\n")
# Compute the recall
nb_recall = recall_score(y_test, y_pred)
print("Naive Bayes recall:", nb_recall)
print("\n")
3
Naïve Bayesian Classifier accuracy: 0.9562043795620438
Naive Bayes precision: 0.9482758620689655
Naive Bayes recall: 0.9482758620689655
[15]: # ANN
from sklearn.neural_network import MLPClassifier
# Train the model
ann = MLPClassifier(hidden_layer_sizes=(100,), max_iter=500)
ann.fit(X_train, y_train)
# Make predictions
4
y_pred = ann.predict(X_test)
# Compute confusion matrix for Naive bayes
sns.heatmap(confusion_matrix(y_test, y_pred), annot=True, cmap='Blues')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion matrix for testing set using ANN')
plt.show()
print("\n")
# Evaluate performance(Accuracy)
ann_accuracy = accuracy_score(y_test, y_pred)
print("ANN accuracy:", ann_accuracy)
print("\n")
# Compute the precision
ann_precision = precision_score(y_test, y_pred)
print("ANN precision:", ann_precision)
print("\n")
# Compute the recall
ann_recall = recall_score(y_test, y_pred)
print("ANN recall:", ann_recall)
print("\n")
5
ANN accuracy: 0.9781021897810219
ANN precision: 0.9661016949152542
ANN recall: 0.9827586206896551
[16]: # KNN
from sklearn.neighbors import KNeighborsClassifier
# Train the model
knn = KNeighborsClassifier(n_neighbors=5)
knn.fit(X_train, y_train)
# Make predictions
6
y_pred = knn.predict(X_test)
# Compute confusion matrix for Naive bayes
sns.heatmap(confusion_matrix(y_test, y_pred), annot=True, cmap='Blues')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion matrix for testing set using ANN')
plt.show()
print("\n")
# Evaluate performance(Accuracy)
knn_accuracy = accuracy_score(y_test, y_pred)
print("KNN accuracy:", knn_accuracy)
print("\n")
# Compute the precision
knn_precision = precision_score(y_test, y_pred)
print("KNN precision:", knn_precision)
print("\n")
# Compute the recall
knn_recall = recall_score(y_test, y_pred)
print("KNN recall:", knn_recall)
print("\n")
C:\Users\SRIRAM\anaconda3\lib\site-
packages\sklearn\neighbors\_classification.py:228: FutureWarning: Unlike other
reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode`
typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will
change: the default value of `keepdims` will become False, the `axis` over which
the statistic is taken will be eliminated, and the value None will no longer be
accepted. Set `keepdims` to True or False to avoid this warning.
mode, _ = stats.mode(_y[neigh_ind, k], axis=1)
7
KNN accuracy: 0.9635036496350365
KNN precision: 0.9818181818181818
KNN recall: 0.9310344827586207
[23]: x = np.arange(3)
y1 = [nb_accuracy,nb_precision,nb_recall]
y2 = [ann_accuracy,ann_precision,ann_recall]
y3 = [knn_accuracy,knn_precision,knn_recall]
# plot data in grouped manner of bar type
plt.bar(x-0.3, y1, width, color='violet')
8
plt.bar(x, y2, width, color='yellow')
plt.bar(x+0.3, y3, width, color='green')
plt.xticks(x, ['Naive Bayes', 'ANN', 'KNN'])
plt.xlabel("Metrics")
plt.ylabel("Values")
plt.legend(["Accuracy","Precision","Recall","F1_score"])
plt.show()
[0 1 2]
[ ]:
