Machine Learning & Deep Learning Practice

I am a Machine Learning practitioner. Here, my mini-projects are based on Kaggle, Github, Udemy online courses, etc.

Storm Prediction

• Storm Prediction as a binary classification
• Artificial Neural Network
• Heatmap of wind, temperature, reflectivity, and relative vorticity data
• Based on ams-ml-python-course at Github

Classification Problem: ROC curves

• Storm Prediction
• Logistic Regression, Random Forest, Naive Bayes, KNN, SVM, XGBoost, and Gradient Boosting
• AUROC and ROC curves for evaluations of the models

Classification Problem: Churn rate prediction

• Encoding categorical data
• Split the data into the training and test set
• Feature Scaling
• Building ANN and predicting the Churn rate of the bank customers

Regression Problem: Solar Radiation prediction

• Heatmap and scatter plots of weather features (temperature, pressure, solar radiation, etc.)
• Linear regression, Random Forest, Support Vector, XGBoost, Gradient Boosting, and ANN
• Evaluations of different regression models

Advanced house price prediction: Kaggle competition

• EDA - Heatmap, boxplot, and scatter plot
• Missing values and outliers
• Random Forest Regression model with GridSearchCV.

Reference

• https://www.kaggle.com/serigne/stacked-regressions-top-4-on-leaderboard
• https://www.kaggle.com/erikbruin/house-prices-lasso-xgboost-and-a-detailed-eda
• https://www.kaggle.com/dejavu23/house-prices-eda-to-ml-beginner