Releases: nkoda/Work-Sample_Data-Engineer
Releases · nkoda/Work-Sample_Data-Engineer
Data Pipeline for Predicting Trading Volume
Description
This release contains a data pipeline that predicts the trading volume of equities using a LightGBM model. The pipeline consists of several components, including data ingestion, data transformation, feature engineering, model training, and a Flask web app for serving predictions.
Key features of the project release include:
- Efficient data ingestion using pandas and thread pools to reduce ingestion time
- Data transformation and feature engineering for preparing the data for modeling
- A LightGBM model trained on the transformed data for predicting trading volume
- A Flask web app for serving predictions to users
- Docker containerization for easy deployment and reproducibility