OluAyo06/House-Pricing

This project focuses on predicting the future prices of houses across selected states in Nigeria. The dataset includes different property types, ranging from bungalows to duplexes. Our goal is to build a reliable model that supports accurate real-estate pricing decisions. --- # **Housing Project** ## **Project Overview** This project focuses on predicting house prices across selected states in Nigeria. Using a structured machine-learning workflow, the notebook walks through data loading, cleaning, exploratory analysis, feature engineering, model training, evaluation, and generating predictions suitable for submission or deployment. The dataset includes various property types such as **bungalows** and **duplexes**, combined with essential housing attributes that support accurate price modelling. --- ## **Notebook Structure & Step-by-Step Explanation** ### **1. Loading of the Dataset** The notebook begins by loading the training and test datasets (CSV format). Initial checks include: * Viewing first few rows * Observing data shape * Checking column names * Previewing target variable (price) ### **2. Information on the Train Dataset** A detailed inspection using: * `df.info()` for data types * `df.describe()` for statistical summary * Checking missing values * Confirming the presence of all required features ### **3. Information on the Test Dataset** The same structural checks are repeated for the test dataset, ensuring alignment with the training data and identifying any mismatched or missing features. ### **4. Cleaning of Dataset** General cleaning performed includes: * Handling missing values (imputation or removal) * Converting data types (numeric, categorical) * Removing duplicates * Fixing inconsistent values ### **5. Cleaning of Train Dataset** Train-specific cleaning ensures the model learns from reliable data. This may involve: * Outlier removal * Feature normalization * Target validation * Dropping irrelevant columns ### **6. Test Dataset Cleaning** Mirrors the transformations applied to the training dataset to maintain consistency during inference. --- ## **Exploratory Data Analysis (EDA)** ### **7. Exploration of the Dataset** Provides an overview of relationships between variables, distribution sha …