search

zxzhijia / Kaggle_TReNDS_Neuroimaging

Apache License 2.0
0 stars 0 forks source link

Study existing tricks/methods people use to get good performance #4

Open zxzhijia opened 4 years ago

zxzhijia commented 4 years ago

https://www.kaggle.com/zxzhijia/model-comparison/edit

zxzhijia commented 4 years ago
  1. LinearRegression
  2. GradientBoostingRegressor: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html
  3. sv regression model: https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVR.html
  4. ridge regression model
zxzhijia commented 4 years ago

SVM gives the lowest so far score in notebooks. 0.159

https://www.kaggle.com/aerdem4/rapids-svm-on-trends-neuroimaging

zxzhijia commented 4 years ago

This guy explained why SVM works pretty good vs NN on this data. It seems cuML is a good thing to learn :)

https://www.kaggle.com/c/trends-assessment-prediction/discussion/146831