app.py

import streamlit as st
import pandas as pd
import numpy as np
import joblib
import textwrap

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.impute import SimpleImputer
from sklearn.feature_selection import SelectKBest, f_classif, f_regression
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, mean_absolute_error, mean_squared_error, r2_score
from imblearn.over_sampling import SMOTE

# Streamlit UI
st.title("🚀 AI Code Generator")
st.markdown("Generate & Train ML Models with Preprocessing and Feature Selection")

# Sidebar UI
st.sidebar.title("Choose Options")
model_options = ["KNN", "SVM", "Random Forest", "Decision Tree", "Perceptron"]
model = st.sidebar.selectbox("Choose a Model:", model_options)

task_options = ["Classification", "Regression"]
task = st.sidebar.selectbox("Choose a Task:", task_options)

# Load Dataset
st.markdown("### Upload your Dataset (CSV)")
uploaded_file = st.file_uploader("Choose a CSV file", type="csv")

if uploaded_file:
    data = pd.read_csv(uploaded_file)
    st.write("Preview of Dataset:", data.head())

    # Preprocessing Steps
    st.markdown("### Data Preprocessing Steps")
    
    # Handling Missing Values
    st.write("✅ Handling missing values using `SimpleImputer`")
    imputer = SimpleImputer(strategy="mean")
    data.fillna(data.mean(), inplace=True)

    # Encoding Categorical Variables
    st.write("✅ Encoding categorical variables")
    for col in data.select_dtypes(include=["object"]).columns:
        data[col] = LabelEncoder().fit_transform(data[col])

    # Splitting Data
    X = data.iloc[:, :-1]  # Features
    y = data.iloc[:, -1]   # Target

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

    # Feature Scaling
    st.write("✅ Applying StandardScaler")
    scaler = StandardScaler()
    X_train = scaler.fit_transform(X_train)
    X_test = scaler.transform(X_test)

    # Handle Imbalanced Dataset using SMOTE
    if task == "Classification":
        st.write("✅ Handling Imbalanced Dataset using SMOTE")
        smote = SMOTE()
        X_train, y_train = smote.fit_resample(X_train, y_train)

    # Feature Selection
    st.write("✅ Selecting Best Features")
    selector = SelectKBest(f_classif if task == "Classification" else f_regression, k=min(5, X.shape[1]))
    X_train = selector.fit_transform(X_train, y_train)
    X_test = selector.transform(X_test)

    # Model Training
    model_mapping = {
        "KNN": "KNeighborsClassifier" if task == "Classification" else "KNeighborsRegressor",
        "SVM": "SVC" if task == "Classification" else "SVR",
        "Random Forest": "RandomForestClassifier" if task == "Classification" else "RandomForestRegressor",
        "Decision Tree": "DecisionTreeClassifier" if task == "Classification" else "DecisionTreeRegressor",
        "Perceptron": "Perceptron" if task == "Classification" else "Perceptron"
    }

    model_class = model_mapping[model]

    template = f"""
import numpy as np
import pandas as pd
import joblib

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.impute import SimpleImputer
from sklearn.feature_selection import SelectKBest, f_classif, f_regression
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, mean_absolute_error, mean_squared_error, r2_score
from imblearn.over_sampling import SMOTE
from sklearn.{model.lower()} import {model_class}

# Load Dataset
data = pd.read_csv('dataset.csv')

# Handling Missing Values
imputer = SimpleImputer(strategy="mean")
data.fillna(data.mean(), inplace=True)

# Encoding Categorical Variables
for col in data.select_dtypes(include=["object"]).columns:
    data[col] = LabelEncoder().fit_transform(data[col])

# Splitting Data
X = data.iloc[:, :-1]
y = data.iloc[:, -1]

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

# Feature Scaling
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

# Handle Imbalanced Data (SMOTE)
if "{task}" == "Classification":
    smote = SMOTE()
    X_train, y_train = smote.fit_resample(X_train, y_train)

# Feature Selection
selector = SelectKBest(f_classif if "{task}" == "Classification" else f_regression, k=min(5, X.shape[1]))
X_train = selector.fit_transform(X_train, y_train)
X_test = selector.transform(X_test)

# Model Training
model = {model_class}()
model.fit(X_train, y_train)

# Save Trained Model
joblib.dump(model, 'models/trained_model.pkl')

# Evaluation Metrics
if "{task}" == "Classification":
    y_pred = model.predict(X_test)
    print("Accuracy:", accuracy_score(y_test, y_pred))
    print("Precision:", precision_score(y_test, y_pred, average='weighted'))
    print("Recall:", recall_score(y_test, y_pred, average='weighted'))
    print("F1 Score:", f1_score(y_test, y_pred, average='weighted'))
else:
    y_pred = model.predict(X_test)
    print("Mean Absolute Error:", mean_absolute_error(y_test, y_pred))
    print("Mean Squared Error:", mean_squared_error(y_test, y_pred))
    print("R2 Score:", r2_score(y_test, y_pred))
"""

    st.code(template, language="python")
    st.download_button("📥 Download AI Model Code", template, "ai_model.py")

    # Save Model
    model_instance = eval(model_class)()
    model_instance.fit(X_train, y_train)
    joblib.dump(model_instance, "models/trained_model.pkl")

    st.success("✅ Model trained and saved as `trained_model.pkl`")