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 · Mini Project
COVID-19 Case Forecasting with LSTM
Forecasts daily case counts with LSTM against classical baselines
1-2 weeks build
Overview
Pulls daily case series from Our World in Data, smooths reporting artefacts, and trains an LSTM to forecast the next one to two weeks. A naive baseline and ARIMA are fitted on the same windows, and the report compares them with MAE and RMSE using walk-forward validation rather than a random split.
What makes it stand up
- 01Case series from the public Our World in Data COVID-19 dataset
- 02Seven-day smoothing to remove weekend reporting dips
- 03LSTM compared with naive and ARIMA baselines
- 04Walk-forward validation with MAE and RMSE per horizon
- 05Country selector and forecast bands in a Streamlit dashboard
Module breakdown
- 01
Data ingestion
Download, country filtering and date alignment.
- 02
Preprocessing
Smoothing, scaling and sliding-window sequences.
- 03
Baselines
Naive persistence and ARIMA fitted per country.
- 04
LSTM model
Stacked LSTM with early stopping and multi-step output.
- 05
Dashboard
Streamlit charts of history, forecast and error.
After this, you will be able to
- Explain how an LSTM carries information across time steps
- Defend walk-forward validation over a random split
- Demo forecasts for a country chosen by the examiner
Technology stack
- Python
- TensorFlow
- Pandas
- statsmodels
- Streamlit
- Level
- Mini Project
- Domain
- Data Science
- Build time
- 1-2 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