1-) Welcome to "Tree Types Prediction Project".
In this project, you must apply EDA processes for the development of predictive models. Handling outliers, domain knowledge and feature engineering (using sqlite3 library) will be challenges.
Also, this project aims to improve your ability to implement algorithms for Multi-Class Classification. Thus, you will have the opportunity to implement many algorithms commonly used for Multi-Class Classification problems.
Before diving into the project, please take a look at the determines and tasks.
NOTE: This project assumes that you already know the basics of coding in Python. You should also be familiar with the theory behind classification models and scikit-learn module as well as Machine Learning before you begin.
2-) We deal with tree type dataset in this notebook.
*1* We loaded the dataset and explored it, order to understand the data.We did Explatonary Data Analysis on the data
*2* We Detect Missing Values and Outliers and removed them.
*3* We focused on numerical and categorical data, Detect Number of Unique values of each column,Focus on Target Variable (Cover_Type)
*4* Detect relationships and correlations between independent variables and target variable.
*5* visualize the class frequencies of the target variable.
*6* Detect relationships and correlations between independent variables.
*7* Consider dropping features that contain little data or that you think will not contribute to the model.
*8* We trained our dataset with various classifier method after clean data from outlier and missing values
*9* And Compare model scores with each other. and selected much suitable model