How mohammed-abu-oudah/predication-of-product-sales works
This repository is a notebook-based product-sales prediction project. Its documented workflow inspects and cleans data, uses plots to explore distributions and relationships, then trains and compares linear regression and random-forest regression; GridSearchCV is used to tune the random forest. The README reports overall weak performance and suggests feature removal or outlier processing as possible improvements. The tree contains only the notebook and README, so the graph uses the notebook as the supported path for each workflow stage; its internal implementation and exact data flow could not be verified from source excerpts.
Main entry points: Prediction_of_Product_Sales.ipynb Inspect and clean
Components 8
Data preparation
- Inspect and clean
Prediction_of_Product_Sales.ipynb
Data exploration
- Explore with plots
Prediction_of_Product_Sales.ipynb
Modeling
- Linear regression
Prediction_of_Product_Sales.ipynb - Random forest
Prediction_of_Product_Sales.ipynb - GridSearchCV tuning
Prediction_of_Product_Sales.ipynb - Model results
Prediction_of_Product_Sales.ipynb
Other components
- Analyst
- Sales data
Connections 10
- Analyst → Inspect and clean: runs
- Sales data → Inspect and clean: input
- Inspect and clean → Explore with plots: explores
- Inspect and clean → Linear regression: trains on
- Inspect and clean → Random forest: trains on
- Random forest → GridSearchCV tuning: tunes with
- Linear regression → Model results: reports
- Random forest → Model results: reports
- GridSearchCV tuning → Model results: reports
- Model results → Analyst: presents