Predicting Sale Price of Items during a Black Friday sale.
The challenge is to predict purchase prices of various products purchased by customers based on historical purchase patterns. The data contained features like age, gender, marital status, categories of products purchased, city demographics etc. Predicting-Black-Friday-Sales.ipynb notebook tackle this problem and has following parts -
-
Part I. Data Ingestion This part involve setting up working directories, importing data, inspecting the data for features, data type of columns, and understanding statistical information from data.
-
Part II. Data Wrangling This part involve curating, cleaning, massaging and creating new features in the data. It also perform imputation in the data for missing or NaN values.
-
Part III. Data Exploration This part help identify significant features, feature correlation, identify features with outliers. This is done by performaing Univariate and Multivariate Analysis. Performed various visualizations like Box-plot, Histogram, Violin Plors, Scatter Plots etc.
-
Part IV. Building Regression Model In this project I build three different models - Linear Regression(Linear Model), Decision Tree(Non-Linear Model) and Gradient Boosting.
-
Part V. Model Evaluation This notebook utilize Root Mean Square Error(RMSE) and Mean Absolute Error(MAE) to evaluate the models created.
-
Part VI. Results This part highlight the results, compare the different models created and in the end generated predicted values for test data.