Mini Project
Data Science
Personal Finance Analytics
Statement parsing, auto-categorisation and spend forecasting
- Python
- Pandas
- Prophet
- Streamlit
- +2
₹2,299₹3,999
4.5/198 reviews
Data Science · Major Project
Weather Forecasting with Machine Learning
Predicts next-day temperature and rain from historical observations
3-4 weeks build
Overview
Forecasts next-day temperature as a regression task and rain or no rain as a classification task, using the public Daily Delhi Climate and Rain in Australia datasets from Kaggle. Lag features, rolling means and seasonal encodings feed random forest, XGBoost and LSTM models, each evaluated with time-ordered splits. A dashboard shows forecasts beside what actually happened.
What makes it stand up
- 01Lag, rolling-window and cyclical date features
- 02Temperature regression and rain classification tracks
- 03Random forest, XGBoost and LSTM compared
- 04Time-ordered validation to avoid look-ahead leakage
- 05Forecast versus actual dashboard with error bands
Module breakdown
- 01
Preparation
Fills gaps and derives lag and rolling features.
- 02
Regression
Predicts next-day temperature with three models.
- 03
Classification
Predicts rain with class-weighted models.
- 04
Evaluation
RMSE, ROC curves and confusion matrices in the report.
- 05
Dashboard
Streamlit charts of forecasts against observations.
After this, you will be able to
- Explain why random splits leak future data in time series
- Defend your feature engineering choices
- Demo forecasts and explain a day the models got wrong
Technology stack
- Python
- Pandas
- scikit-learn
- XGBoost
- TensorFlow
- Streamlit
- Level
- Major Project
- Domain
- Data Science
- Build time
- 3-4 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support