kaggle:PUBG Finish Placement Prediction

The Mission of Machine Learning :PUBG Finish Placement Prediction

一. Introduction
二. Experiments
三. Conclude

Ⅰ. Introduction

Based on the background of PlayerUnknown’s BattleGrounds (PUBG), we can use over 65,000 games’ worth of anonymized player data, split into training and testing sets, and asked to predict final placement from final in-game stats and initial player ratings, where 0 is last place and 1 is winner , chicken dinner.

Ⅱ. Experiments

After going through a series of groping, drawing and submission, I can initially understand what is going on.

1 . First day of trying to communicate with Kaggle

2 . Second day of trying to communicate with Kaggle

3 . The software of JupyterLab is useful to debug with python，it could run step by step, and you can find which step is in the wrong direction.

4 . Although this is my first feature engineering analysis, with the help of notebooks, it is still a start. (please ignore my poor English).

5 . After working and debugging, some results were received.

Ⅲ.Conclude

First of all, I am fortunate to be able to join the course in machine learning. I am grateful to the instructors for their examples and guidance in the classroom. Actually, I have no idea about machine learning before this course and just know that it is popular and useful for classification problems. Learning through the course and this mission, I think I have taken the first step of machine learning and I also have a long way to learn deeply.