You can download the dataset here.
Description about the dataset:

• The data has many features
• Half of the features are category
• The number of instance is small (<2000)
• Many features are rate-based category (Good/Average/Bad)

About my submission:

• You can track my work in main.ipynb
• Spend most of the time for Data Analyst
• First time using sns to plot scatter and boxplot
• Involved in handling many DataFrame cols
• Implement data selection for both num and cate features
• Effectively using corr_matrix to choose features
• First time implemente ExtraTree to extract features
• Combined both early stoping and SearchCv to fine tune

about my note:

• learn more about some new plots
• some new definitions in statistic
• scipy

about my remade:

• apply Graident Boosting
• skip the data analysis