Commits · dcbb16f27d62d2798b9d9ebbb42af29480cb5008 · Luc Giffon / bolsonaro · GitLab

Snippets Groups Projects

Mar 29, 2020
- Integrate non negative omp versions in the experiment pipeline · dcbb16f2
  Charly Lamothe authored 5 years ago
  
  dcbb16f2
Mar 28, 2020
- Started to think on how to integrate omp_nn to the training pipeline · d65fd5bb
  Charly Lamothe authored 5 years ago
  
  d65fd5bb
Mar 27, 2020
- Few fixes, metrics, viz (TODO: clean compute_results.py :x) · 0ea777e7
  Charly Lamothe authored 5 years ago
  
  0ea777e7
Mar 25, 2020
- Fix flw_pairs loading. Prepare all new exps: omp_distillation, preds... · 87fb436c
  Charly Lamothe authored 5 years ago
  87fb436c
Mar 24, 2020
- fix and optimise ensemble selection forest regressor · 8d52496f
  Luc Giffon authored 5 years ago
  
  8d52496f
- Handle similarity_similarities and similarity_predictions in the pipeline and... · 5ee9422b
  Charly Lamothe authored 5 years ago
  5ee9422b
Mar 13, 2020
- Merge from master · 4d4c0848
  Charly Lamothe authored 5 years ago
  
  4d4c0848
Mar 12, 2020
- Fix on resume mode for extracted forest size job · 9d7ef0e7
  Charly Lamothe authored 5 years ago
  
  9d7ef0e7
Mar 06, 2020
- Correction on random extraction · 6483c0dc
  Léo Bouscarrat authored 5 years ago
  
  6483c0dc
- Update fix random strategy (wip · 138660cb
  Charly Lamothe authored 5 years ago
  
  138660cb
- Fix hyperparams bugs in base and random. Fix extracted forest size used in... · 2a265135
  Charly Lamothe authored 5 years ago
  2a265135
- Integrate Paolo's code of method 'Ensemble selection from libraries of models'... · 1194ee2f
  Charly Lamothe authored 5 years ago
  1194ee2f
- Speedup similarity forest regressor and add parallelization at the extracted... · 29a11860
  Charly Lamothe authored 5 years ago
  29a11860
- Fix parallelization and estimator default hyperparams in kmeans and similarity... · 0a97ff64
  Charly Lamothe authored 5 years ago
  0a97ff64
- Add on resume mode for the experiment training (and set the overwrite of the... · be5bc24a
  Charly Lamothe authored 5 years ago
  be5bc24a
Mar 05, 2020
- Changing endpoint linspace · 006b001e
  Léo Bouscarrat authored 5 years ago
  
  006b001e
Feb 28, 2020
- Finish to add similarity method to the pipeline. Add kmeans pruning method · 59e65276
  Charly Lamothe authored 5 years ago
  
  59e65276
Feb 04, 2020
- Add Paolo's first implementation of this paper:... · c80ddd61
  Charly Lamothe authored 5 years ago
  c80ddd61
Jan 09, 2020
- - Add a temp fix for the subset used in base and random strategies; · 21ccc627
  Charly Lamothe authored 5 years ago
  21ccc627
Jan 08, 2020
- - Add diamonds results; · 9ce94798
  Charly Lamothe authored 5 years ago
  9ce94798
Dec 29, 2019
- - Remove extracted_forest_sizes_number parameter from compute_results.py and... · 559d73c0
  Charly Lamothe authored 5 years ago
  559d73c0
Dec 26, 2019
- - Add code for stages 2 and 3 results; · 8de5e96a
  Charly Lamothe authored 5 years ago
  8de5e96a
- - Add command lines for stage2 experiments; · 58061ea4
  Charly Lamothe authored 5 years ago
  58061ea4
Dec 20, 2019
- - Unignore results; · 51ba8a0e
  Charly Lamothe authored 5 years ago
  51ba8a0e
Dec 19, 2019
- - Finish first stable version of stage1 plotting; · 649b4e64
  Charly Lamothe authored 5 years ago
  649b4e64
- - Reduce the extracted forest sizes upper bound and number because OMP seems... · 6a6cf747
  Charly Lamothe authored 5 years ago
  6a6cf747
Dec 18, 2019
- POC of possible wrong way to compute best hyperparams. Are there the best only... · fd6dbc7b
  Charly Lamothe authored 5 years ago
  fd6dbc7b
- - Add an option to not use the best hyperparameters file; · 880ff78f
  Charly Lamothe authored 5 years ago
  880ff78f
- - Replace the futures concurrence by joblib Parallel (and add optional tqdm progress bar); · 11017545
  Charly Lamothe authored 5 years ago
  11017545
Dec 01, 2019
- - Add standard dataset scaling for dataset normalization; · f61314f3
  Charly Lamothe authored 5 years ago
  f61314f3
Nov 22, 2019
- - Split train function in three distinct functions; · a0f7c96f
  Charly Lamothe authored 5 years ago
  a0f7c96f
- When training, look if there is bayesian search results, if yes use this.... · c66d117d
  Léo Bouscarrat authored 5 years ago
  c66d117d
- add multiclass classifier mais attention ya un bug dans le calcul du score · 5e50bbaa
  Luc Giffon authored 5 years ago
  
  5e50bbaa
Nov 21, 2019
- Big changes: Create intermediate classes OMPForest and SingleOmpForest for... · 3f5cdf68
  Luc Giffon authored 5 years ago
  3f5cdf68
Nov 20, 2019
- Add functions to do bayesian hyperparameters search · bf5803b6
  Léo Bouscarrat authored 5 years ago
  
  bf5803b6
Nov 09, 2019
- Fix a spelling mistake in train.py. · 211dc83a
  Charly LAMOTHE authored 5 years ago
  
  211dc83a
- - Add experiment_configuration parameter to run an experiment from a json... · 789a11a6
  Charly LAMOTHE authored 5 years ago
  789a11a6
- - Compute each computations of a given seed in a dedicated job; · cb0030d8
  Charly LAMOTHE authored 5 years ago
  cb0030d8
Nov 08, 2019
- Add the weights normalization parameter (but not implemented yet) · 0fce0319
  Charly LAMOTHE authored 5 years ago
  
  0fce0319
Nov 06, 2019
- Replace use_dev_subset by subsets_used parameter, in order to specify more... · 7455fd98
  Charly LAMOTHE authored 5 years ago
  7455fd98