app.py

import streamlit as st
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

# Function to build and evaluate models
def build_and_evaluate_model(model, X_train, X_test, y_train, y_test):
    try:
        model.fit(X_train, y_train)
        y_pred = model.predict(X_test)
        # Convert y_pred to the same data type as y_test for comparison
        y_pred = y_pred.astype(str)  #Used in data manipulation
        accuracy = accuracy_score(y_test.astype(str), y_pred)
        return accuracy
    except Exception as e:
        return f"Not Applicable: {str(e)}"

# Streamlit app
def main():
        # Custom CSS for styling the title
    custom_css = """
        <style>
            .title-text {
                font-size: 55px;
                font-weight: bold;
                color: #FF0A01; /* Dark Blue */
                text-align: center;
                margin-bottom: 0px;
            }
            .tagline-text {
                font-size: 18px;
                font-style: italic;
                color: #2c3e50; /* White */
                text-align: center;
                margin-top:-10px;
                margin-bottom: 20px;
            }
        </style>
    """

    # Streamlit app
    st.markdown(custom_css, unsafe_allow_html=True)
    # Title and tagline
    st.markdown("<p class='title-text'>ModelMetrics Hub</p>", unsafe_allow_html=True)
    st.markdown("<p class='tagline-text'>Optimal Model Explorer</p>", unsafe_allow_html=True)

    #st.title("ModelMetrics Hub \n _Optimal Model Explorer_")
    st.sidebar.title("Model Selection")
    model_names = ["Random Forest", "Decision Tree", "KNN", "SVM", "Logistic Regression"]
    selected_models = st.sidebar.multiselect("Choose models to compare", model_names, default=model_names)

    uploaded_file = st.file_uploader("Upload a CSV file", type=["csv"])
    if uploaded_file is not None:
        data = pd.read_csv(uploaded_file, encoding='latin-1')  # Specify encoding here
        st.write("Preview of the dataset:")
        st.write(data.head())

        feature_columns = st.multiselect("Select feature columns", data.columns.tolist())
        target_column = st.selectbox("Select target column", data.columns.tolist())

        if st.button("Run Models"):
            X = data[feature_columns]
            y = data[target_column]

            y = y.astype(str)

            # Perform one-hot encoding for categorical columns
            categorical_cols = X.select_dtypes(include=['object']).columns.tolist()
            X = pd.get_dummies(X, columns=categorical_cols, drop_first=True)

            X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

            models = {
                "Random Forest": RandomForestClassifier(n_estimators=100, random_state=42),
                "Decision Tree": DecisionTreeClassifier(random_state=42),
                "KNN": KNeighborsClassifier(n_neighbors=5),
                "SVM": SVC(kernel='rbf', random_state=42),
                "Logistic Regression": LogisticRegression(max_iter=1000, random_state=42)
            }

            results = {}
            for model_name in selected_models:
                accuracy = build_and_evaluate_model(models[model_name], X_train, X_test, y_train, y_test)
                results[model_name] = accuracy

            sorted_results = sorted(results.items(), key=lambda x: x[1], reverse=True)  #key=lambda x: is a custom sorting order 

            st.subheader("Accuracy")
            for model_name, accuracy in sorted_results:
                st.write(f"{model_name}: {accuracy * 100:.2f}%")

if __name__ == "__main__":
    main()