Kaggle Notebooks

• Predictive modeling Notebooks for Kaggle competition
• Tools
  • Language: Python3.7
  • Modeling Framework: Scikit-Learn, PyTorch
  • Hyper-parameter Optimization Tool: Ray Tune
  • Working Environment: Colab Notebook

Tabular Data

Competition / Data	Problem Type	Code
Titanic Survival Prediction	Binary Classification	Notebook(kor)
House Price Prediction	Regression	Notebook(kor)
Forest Cover Type Prediction	Multi-class Classification	Notebook(kor)
Bike Sharing Demand	Multi-output Regression	Notebook(kor)
Prudential Life Insurance Assessment	Ordinal Classification	Notebook(kor)

Sequential Data

Competition / Data	Problem Type	Model	Code
NASA Turbofan Jet Engine	Multivariate Time-Series Scalar Forecasting (Remaining Useful Life Prediction)	LSTM	Notebook
Pump Sensor	Multivariate Time-Series Anomaly Detection	LSTM_AutoEncoder	Notebook

Image Data

Competition / Data	Problem Type	Model	Code
Dog Breed Identification	Classification	ResNet34(Pretrained) ResNet18 NiN	Notebook Notebook Notebook
Generative Dog Images	Generation	DCGAN	Notebook
I’m Something of a Painter Myself	Style Transfer	Neural Style Transfer	Notebook
PASCAL VOC 2012	Semantic Segmentation	FCN_ResNet18	Notebook