Commits · 31d12659ce085f9433b79cc83dceae8d125eb841 · Luc Giffon / bolsonaro · GitLab

Dec 01, 2019
- Fix few typos in new compute_hyperparams version. · 31d12659
  Charly Lamothe authored Dec 1, 2019
  
  31d12659
- Remove useless Real import · 39d4540f
  Charly Lamothe authored Dec 1, 2019
  
  39d4540f
- Rename 'use_different_seed_number' parameter to 'use_variable_seed_number' · 7494dbc4
  Charly Lamothe authored Dec 1, 2019
  
  7494dbc4
- Compute best hyperparameters over k seeds, and select the most commons. · e319343d
  Charly Lamothe authored Dec 1, 2019
  
  e319343d
- By default, compute the hyperparams of the whole dataset list. · 254e846e
  Charly Lamothe authored Dec 1, 2019
  
  254e846e
- - Add standard dataset scaling for dataset normalization; · f61314f3
  Charly Lamothe authored Dec 1, 2019
  f61314f3
- Finish to fix sklearn dataset fetchers. Keep 20newsgroups_vectorized instead... · 967742a6
  Charly Lamothe authored Dec 1, 2019
  967742a6
Nov 29, 2019
- Fix most of the missing dataset fetchers. Remainings: 20newsgroups,... · d70fad15
  Charly Lamothe authored Nov 29, 2019
  d70fad15
Nov 22, 2019
- - Split train function in three distinct functions; · a0f7c96f
  Charly Lamothe authored Nov 22, 2019
  a0f7c96f
- Clean up some imports · f866e30d
  Charly Lamothe authored Nov 22, 2019
  
  f866e30d
- - Add missing LOG_PATH in two init files; · a86dff87
  Charly Lamothe authored Nov 22, 2019
  a86dff87
- When training, look if there is bayesian search results, if yes use this.... · c66d117d
  Léo Bouscarrat authored Nov 22, 2019
  c66d117d
- solve bug scoring multiclas · 7bb11288
  Luc Giffon authored Nov 22, 2019
  
  7bb11288
- add multiclass classifier mais attention ya un bug dans le calcul du score · 5e50bbaa
  Luc Giffon authored Nov 22, 2019
  
  5e50bbaa
Nov 21, 2019
- add experiments directory and print score on train and dev · f7e772d9
  Luc Giffon authored Nov 21, 2019
  
  f7e772d9
- Big changes: Create intermediate classes OMPForest and SingleOmpForest for... · 3f5cdf68
  Luc Giffon authored Nov 21, 2019
  3f5cdf68
Nov 20, 2019
- Add functions to do bayesian hyperparameters search · bf5803b6
  Léo Bouscarrat authored Nov 20, 2019
  
  bf5803b6
Nov 09, 2019
- Introduce the notion of "stage" in the experiments (see TODO comment in... · 9a9a3bff
  Charly LAMOTHE authored Nov 9, 2019
  9a9a3bff
- Add TODO list of the figure scheme in compute_results.py · 9830bbe0
  Charly LAMOTHE authored Nov 9, 2019
  
  9830bbe0
- Fix a spelling mistake in train.py. · 211dc83a
  Charly LAMOTHE authored Nov 9, 2019
  
  211dc83a
- - Add experiment_configuration parameter to run an experiment from a json... · 789a11a6
  Charly LAMOTHE authored Nov 9, 2019
  789a11a6
- - Compute each computations of a given seed in a dedicated job; · cb0030d8
  Charly LAMOTHE authored Nov 9, 2019
  cb0030d8
Nov 08, 2019
- Fix normalize_weights impl · 7dd2aab5
  Charly LAMOTHE authored Nov 8, 2019
  
  7dd2aab5
- Add comments in weight_density function and faster the zero trimming · 708b1731
  Charly LAMOTHE authored Nov 8, 2019
  
  708b1731
- Use OMP to compute the prediction instead of doing it manually (same results) · 8b3a6c49
  Charly LAMOTHE authored Nov 8, 2019
  
  8b3a6c49
- Add the weights normalization parameter (but not implemented yet) · 0fce0319
  Charly LAMOTHE authored Nov 8, 2019
  
  0fce0319
Nov 06, 2019
- Replace use_dev_subset by subsets_used parameter, in order to specify more... · 7455fd98
  Charly LAMOTHE authored Nov 6, 2019
  7455fd98
- - Fix weight_density. TODO: add legend (colored by seed) · eefd7b6b
  Charly LAMOTHE authored Nov 6, 2019
  View commits for tag archive/5-add-plots-2 archive/5-add-plots-2
  
  eefd7b6b
- Add elements on requirements.txt · 4a5324ba
  Léo Bouscarrat authored Nov 6, 2019
  
  4a5324ba
Nov 05, 2019
- Remove two useless concatenate in trainer.py to save space and time · d89d9d52
  Charly LAMOTHE authored Nov 5, 2019
  
  d89d9d52
- Add some comments in plot_losses · d67e49c8
  Charly LAMOTHE authored Nov 5, 2019
  
  d67e49c8
- Add comments on compute_results.py file · fc19c39e
  Charly LAMOTHE authored Nov 5, 2019
  
  fc19c39e
- Train the forest on train and OMP on dev OR train both the forest and OMP on train+dev · 28b804c6
  Charly LAMOTHE authored Nov 5, 2019
  
  28b804c6
- Add an option to disable the progress bars · 9199d9bb
  Charly LAMOTHE authored Nov 5, 2019
  
  9199d9bb
- Uppercased default consts · 306656fc
  Charly LAMOTHE authored Nov 5, 2019
  
  306656fc
- Remove useless sorting in resolve_experiment_id · e0abe67e
  Charly LAMOTHE authored Nov 5, 2019
  
  e0abe67e
- Save all attributes of model_raw_results automatically · a6a476ce
  Charly LAMOTHE authored Nov 5, 2019
  
  a6a476ce
- - Remove useless error handling files; · ec569ad5
  Charly LAMOTHE authored Nov 5, 2019
  ec569ad5
- Remove useless files visualize.py for now · 44e43b58
  Charly LAMOTHE authored Nov 5, 2019
  
  44e43b58
- - Finish plot_losses implementation; · 5e990b2b
  Charly LAMOTHE authored Nov 5, 2019
  5e990b2b