Parth Pawar
Engineer · Builder · Entrepreneur
← All work

Machine learning · Streamlit

LibraIQ

A demand classifier for university library books. Compares models and presents High, Medium, or Low predictions in an interactive dashboard.

0

demand classes

multi

models compared

live

Streamlit dashboard

The demand problem

A university library has to decide which books need more copies and which sit on the shelf. Without a prediction, those decisions are guesswork.

One dashboard

An interactive dashboard where a book's demand comes back as High, Medium, or Low, backed by a model chosen through comparison rather than assumption.

The pipeline

01

Prepared the library data with Pandas for a classification task.

02

Framed demand as three classes: High, Medium, and Low.

03

Trained and compared several Scikit-learn models instead of committing to one.

04

Chose the model from that comparison.

05

Wrapped the predictions in an interactive Streamlit dashboard.

On the board

  • Book demand classification

  • Model comparison with Scikit-learn

  • High / Medium / Low predictions

  • Pandas data preparation

  • Interactive Streamlit dashboard

  • Open repository