raghava-110/Ham_or_Spam_classifier

1import streamlit as st2import pandas as pd3import numpy as np4import sklearn5from sklearn.model_selection import train_test_split6from sklearn.feature_extraction.text import CountVectorizer7from sklearn.neighbors import KNeighborsClassifier8from sklearn.naive_bayes import MultinomialNB9from sklearn.tree import DecisionTreeClassifier10from sklearn.linear_model import LogisticRegression11from sklearn.svm import SVC12from sklearn.metrics import accuracy_score13 14# Load Dataset15df = pd.read_csv("spam.csv")16 17# Title18st.title(":blue[ Spam and Ham Emails Using ML Algorithms]")19 20# Preparing Data21x = df["Message"]22y = df["Category"]23 24bow = CountVectorizer(stop_words="english")25final_data = pd.DataFrame(bow.fit_transform(x).toarray(), columns=bow.get_feature_names_out())26 27x_train, x_test, y_train, y_test = train_test_split(final_data, y, test_size=0.2, random_state=20)28 29# Available Models30models = {31    "Naive Bayes": MultinomialNB(),32    "KNN": KNeighborsClassifier(),33    "Decision Tree": DecisionTreeClassifier(),34    "Logistic Regression": LogisticRegression(),35    "SVM": SVC()36}37 38# Select Model39model_choice = st.selectbox("Choose a Classification Algorithm", list(models.keys()))40 41# Train and Evaluate Model42obj = models[model_choice]43obj.fit(x_train, y_train)44y_pred = obj.predict(x_test)45accuracy = accuracy_score(y_test, y_pred)46 47# Show Accuracy when button is clicked48if st.button("Show Accuracy"):49    st.write(f"**Accuracy of {model_choice}:** {accuracy:.4f}")50 51# Input Field for Email52email_input = st.text_input("enter email")53 54# Prediction Function55def predict_email(email):56    data = bow.transform([email]).toarray()  # Convert sparse to dense57    prediction = obj.predict(data)[0]58    st.write(f"**Prediction:** {prediction}")59 60# Predict Button61if st.button("Predict Email"):62    if email_input:63        predict_email(email_input)64    else:65        st.write(":blue[enter mail]")66 67