Оригинал тут → https://github.com/Data-Learn/data-science/blob/main/ML-101%20Modules/Module%2001/Lesson%2003/Plan.ipynb

С курса по ML от Анастасии Рицца на datalearn.ru

Общее название проекта

«Детальное название проекта.»

Table of Contents

Part 0: Introduction

Overview

О чем этот датасет
+
Метаданные:

• Rank — Ranking of overall sales
• Name — The games name
• Year — Year of the game’s release

Assumptions

Пояснения/уточнения

Questions:

 Вопросы но котороые надо ответить

• #### Question 1:
• #### Question 2:
• #### Question 3:

Part 1: Import, Settings, Load Data

• ### Import libraries, Create settings, Read data from ‘.csv’ file

Part 2: Exploratory Data Analysis

• ### Info, Head, Describe
• ### Observation of target variable «…»
• ### Missing Data
  • #### List of data features with missing values (visualisation: какую диаграмму, график или плот используем?)
  • #### Filling missing values
• ### Numerical and Categorical features
  • #### List of Numerical and Categorical features
  • #### Numerical features:
    • Head
    • Visualisation of Numerical features (какую диаграмму, график или плот используем?)
    • Outliers (visualisation: какую диаграмму, график или плот используем?)
    • Correlation Numerical features to the target
  • #### Categorical Features:
    • Head
    • Visualisation of Categorical features (какую диаграмму, график или плот используем?)
    • Convert Categorical into Numerical features
    • Drop all old Categorical features
  • #### Correlation new features to the target. Drop all features with weak correlation to the target.
  • #### Visualisation of all data features with strong correlation to target (visualisation: heatmap)

Part 3: Data Wrangling and Transformation

• ### Multicollinearity
• ### Standard Scaler
• ### Creating datasets for ML part
• ### ‘Train\Test’ splitting method

Part 4: Machine Learning

• ### ML Models
• ### Build and train a models
• ### Evaluate a models
  • #### If regression: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Abolute Error (MAE), R Squared
  • #### If Classification: Classification Report and Confusion Matrix
• ### Hyper parameters tuning (если надо)
• ### Creating final predictions with Test set
• ### If Classification: AUC–ROC curve (если надо)

Conclusion

• ### Submission of ‘.csv’ file with predictions