Skip to content

Latest commit

 

History

History
53 lines (36 loc) · 1.66 KB

LOG.md

File metadata and controls

53 lines (36 loc) · 1.66 KB
Raw

LOG

Jan, 30, 2018

GossipCat 0.1.40

  • Add bayesOot to the core as a function to tune hyper parameters.

Dec 15, 2017

GossipCat 0.1.32

  • Split Feature from core as a class. Automatical feature engineering on technical level.
  • Split Glimpse from core.
  • Polished SimAnneal, core.

Dec 14, 2017

GossipCat 0.1.24

  • Split Report from core as a class.
  • Polished SimAnneal.

Dec 13, 2017

GossipCat 0.1.10

Tested and modified functions. Add param_1 as default params for simAnneal function.

Dec 12, 2017

gossipcat.py

Add several new functions to do feature engineering:

  • features_dup: This function checks first n_head rows and obtains duplicated features.
  • features_clf: This function divides features into sublists according to their data type.
  • corr_pairs: This function computes correlated feature pairs with correlated coefficient larger than gamma.
  • features_new: This function builds new features based on correlated feature pairs if the new feature has an AUC greater than auc_score with logistic regression.

Dec 11, 2017

gossipcat.py

  • glimpse: This function prints a general infomation of the dataset and plot the distribution of target value.
  • simAnneal: This function uses the simulated annealing to find the optimal hyper parameters and return an optimized classifier.
  • report: This function prints model report with a classifier on test dataset.
  • report_CM: This function prints the recall rate of the classifier on test data and plots out confusion matrix.
  • report_PR: This function plots precision-recall curve and gives average precision.