Commit Graph

367 Commits (78807b941ca35e384f863c2238c1932b320e9dce)

Author SHA1 Message Date
junos 78807b941c Add analysis for composite score of stress. 2023-05-31 21:00:55 +02:00
junos a9af113c9c Add confusion matrices for all methods. 2023-05-31 17:41:50 +02:00
junos 97113fe9ab Sum up confusion matrix and illustrate use with dummy. 2023-05-31 17:27:49 +02:00
junos bc78a1d498 Define a confusion matrix scorer. 2023-05-31 16:05:39 +02:00
junos aca84b214d Small corrections. 2023-05-19 03:19:42 +02:00
junos aa13123136 Handle clustering classification the same as other classification models again. 2023-05-19 03:04:09 +02:00
junos c51e0da0f7 Handle clustering classification the same as other classification models. 2023-05-19 02:52:56 +02:00
junos a2401b5e36 Add multiclass scoring. 2023-05-19 01:34:34 +02:00
junos 70232949c3 Better handling of "medium" category. 2023-05-19 01:10:30 +02:00
junos 8a9595c615 Better handling of input filename again. 2023-05-18 22:53:58 +02:00
junos 045b9fa0dc Better handling of input filename. 2023-05-18 19:03:53 +02:00
junos bb4445f1a8 Add bins to output filename. 2023-05-18 18:58:19 +02:00
junos 1318ae3609 Rename methods to make them consistent with regression methods. 2023-05-18 18:55:31 +02:00
junos 45441c288d Correct column name. 2023-05-18 18:47:56 +02:00
junos fa45b30955 Set output path programmatically. 2023-05-18 18:40:54 +02:00
junos 2336edffb6 Retain metric names in final scores. 2023-05-18 18:40:06 +02:00
junos b756ed5feb Set more parameters as user-specified constants. 2023-05-18 18:06:32 +02:00
junos cad28c3fe8 Set path programmatically. 2023-05-18 16:36:46 +02:00
junos 38a405d378 Add index when inserting one row. 2023-05-17 18:13:20 +02:00
junos 2c5a0b4157 Label plot axes. 2023-05-17 16:32:27 +02:00
junos 0409c9e982 Fix format specification. 2023-05-16 17:22:09 +02:00
junos a7446cc34a Specify columns to aggregate and save figures as pdfs. 2023-05-16 17:05:43 +02:00
junos 118e686491 Specify format directly as infer_datetime_format was deprecated. 2023-05-16 17:04:48 +02:00
junos 9417a1b9f1 Do not break markdown lines. 2023-05-16 16:37:34 +02:00
junos 7b5db88f1d Remove and ignore results. 2023-05-16 16:22:29 +02:00
junos 0f8f0b0fb6 Update URL call. 2023-05-16 16:17:53 +02:00
junos 26c7d22b83 Add an option to save figures. 2023-05-16 16:17:06 +02:00
junos 87781840d4 Use concat instead of append which was deprecated. 2023-05-12 16:32:08 +02:00
junos 3091328fc5 Format comments. 2023-05-11 16:51:38 +02:00
junos 055e87dbac Return scores for classification. 2023-05-10 23:51:12 +02:00
junos f58d20ffc2 Update classification runner. 2023-05-10 23:17:44 +02:00
junos 075fdab9ea Select segment and save results. 2023-05-10 23:00:03 +02:00
junos 91e7352480 Thoroughly refactor classification runner. 2023-05-10 22:50:00 +02:00
junos 35c09374dd Free up memory during model building. 2023-05-10 21:44:40 +02:00
junos b505fb2b6a Thoroughly refactor regression runner. 2023-05-10 20:30:51 +02:00
junos 47b1ecdbb9 First format with black and then check with flake8. 2023-05-10 15:29:32 +02:00
junos 24744c288d Extract one step of preparation into a separate function. 2023-05-10 15:28:09 +02:00
junos caeaf03239 Provide data instead of csv input. 2023-05-10 15:20:33 +02:00
junos cd5d8b6a10 Update rapids and add regex=True.
Reformat debug_heatmap.
2023-05-10 15:12:27 +02:00
junos 3e38b64b45 Merge branch 'ml_pipeline' 2023-05-10 15:02:17 +02:00
junos 76071fd550 Start using pre-commit hooks. 2023-04-24 15:38:54 +02:00
junos c66e046014 Use methods in helper.py. 2023-04-21 21:41:00 +02:00
junos 48118f125d Reformat ml_pipeline_regression.py 2023-04-21 21:34:54 +02:00
junos 583ee82e80 Add xgboost to dependencies and reformat helper.py. 2023-04-21 21:33:06 +02:00
junos 59552c18a9 Update python to 3.11. 2023-04-21 18:08:54 +02:00
Primoz 26804cf8ea Repair preprocessing one hot encoding of test set. 2023-04-21 13:24:31 +02:00
Primoz 865225994b Added testing section after feature selection. 2023-04-20 13:29:14 +02:00
Primoz 259be708aa Improve the feature selection method with validations etc. 2023-04-20 13:26:20 +02:00
Primoz 0594993133 Add GroupKFold to feature selection CV. Start with generic metric calculation procedure. 2023-04-20 11:20:26 +02:00
Primoz 1cbc743cf7 Add kBest method to initially filter out the worst performing features. Update comments. 2023-04-20 10:12:16 +02:00