Skip to content

VegetableDogKing/baldman_final

Repository files navigation

Danceability Estimation

To-Do List

  1. 加權
  2. 刪離群值
  3. transformation
  4. encoding 改一下
  5. desciption

Report 架構

  1. 三種方法: Linear Regression(baseline), Softmax, XGboost, RandomForest??
  2. 選擇feature方法: 相關係數、PCA、DAE
  3. evaluation: cross validation score

更新動態

2023/5/4

  • 測試 n_estimator = 2000, max_depth = 6

2023/5/5

  • 把preprocess新增到util裡面,讓code比較好看
  • 把檔名修改放到前面,從上面改就可以了
  • 可以把GBDT改成 from sklearn.ensemble import GradientBoostingRegressor,這樣就不會是整數,但是參數還要在調一下
  • to-do: 幫我寫一下sklearn.kfold/cross-validate來分資料(直接寫在util.py裡面就可以了),現在還是用train 75% test 25%。應該要分成10個fold然後每個參數跑10次,可以放著讓她跑,記得把每個參數的MAE存下來之後好畫圖

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages