118 Commits

Author SHA1 Message Date
382ae921ab Second try, finished? 2020-05-18 12:55:00 +02:00
e52cbbb192 First approach to Platt scaling 2020-05-18 11:51:27 +02:00
86a9ef2f3a update travis config to notify by mail failures 2020-05-17 19:10:42 +02:00
3cd004350d Fix failing tests 2020-05-17 18:55:33 +02:00
91af875062 Fix issues in notebooks 2020-05-17 18:36:28 +02:00
9f30627e47 Fix main and small issues in notebook test 2020-05-17 18:06:03 +02:00
d997bc3460 Add model comparison to notebook test 2020-05-17 16:17:57 +02:00
ad51d0b36b Update readme, add travis config 2020-05-17 00:13:29 +02:00
ebae3cb316 updt jupyter notebook to fix download of dataset 2020-05-16 22:21:40 +02:00
6a8ef288ce add _linear_function to Stree
First try to implement predict_proba with sigmoid based on Platt
2020-05-16 20:34:03 +02:00
80b5cf8e72 Implement predict & predict_proba optimization
reduces time in two orders of magnitude in creditcard dataset
2020-05-15 23:35:33 +02:00
e56b955b92 Implement predict_proba with test.
Fix tree overload with dataset in nodes only needed in tests
2020-05-14 18:42:17 +02:00
e3ae3a3a6c Add C param in constructor and creditcard dataset 2020-05-14 10:48:39 +02:00
8f71eeb316 Implement predict and score methods & tests 2020-05-13 12:42:09 +02:00
c4de782a3f compute predictor and store model in node 2020-05-13 00:12:05 +02:00
371257c121 Implement split data with or without using predictions & some tests 2020-05-12 17:40:06 +02:00
b9ea24696f Working tree with samples and first test 2020-05-10 14:25:58 +02:00
Ricardo Montañana Gómez
2ca25fd656 Initial commit 2020-05-09 23:01:30 +02:00