The Calories Burnt Prediction project uses machine learning to estimate calories burned during exercise based on personal and workout information. It combines data preprocessing, visualization, feature engineering, model training, evaluation, and cross-validation. An XGBoost regression model predicts calorie expenditure with high accuracy, making it useful for fitness and health applications.
Collects exercise and calorie datasets. Merges both datasets using User_ID. Cleans the data by checking missing values and duplicates. Performs exploratory data analysis using charts. Finds relationships between numerical features with a correlation heatmap. Encodes categorical data (Gender) using OrdinalEncoder. Scales numerical features using StandardScaler. Builds a preprocessing and prediction pipeline using Scikit-learn Pipeline. Trains an XGBoost Regressor to predict calories burned. Evaluates the model using R² Score, Mean Absolute Error (MAE), and 5-Fold Cross Validation for reliable performance.
preprocessor = ColumnTransformer(transformers=[
('ordinal', OrdinalEncoder(), ['Gender']),
('num', StandardScaler(), ['Age', 'Height', 'Weight', 'Duration','Heart_Rate', 'Body_Temp'])
], remainder='passthrough')