Kaggle-FSU23Spring: Housing Prices Prediction

Overview

• overview.ipynb: includes concise introduction of preprocessing data, performing feature engineering and modeling.
• eda_v1.ipynb: contains preprocessing details and first try.
• variables.txt: contains useful variables.

Hyperparameters Tuning (scripts -> results):

• random_forest_tune.py -> rf.csv
• gbm_tune.py -> gbm.csv

Each of them contains 50 trials, and search spaces are included in the overview.ipynb

Useful resources:

• Yggdrasil backend: https://dl.acm.org/doi/10.1145/3580305.3599933#sec-supp
• Model hyperparameters:
  • https://www.tensorflow.org/decision_forests/api_docs/python/tfdf/keras/RandomForestModel
  • https://www.tensorflow.org/decision_forests/api_docs/python/tfdf/keras/GradientBoostedTreesModel