clean scripts

@leo.bouscarrat @luc.giffon

Il reste encore à ajouter plus de plots dans compute_results.py (pour l'instant il y a un exemple qui plot les loss sur train/dev/test).

Question : Dans le notebook il y a une version de la prédiction calculée avec une division de la somme des poids, et une version sans. C'était laquelle la bonne déjà ?

EDIT 1 - Exemple d'utilisation :

python code/train.py --dataset_name=california_housing --normalize_D --dataset_normalizer=standard --extracted_forest_size 10 20 30 --random_seed_number=2
python code/compute_results.py --experiment_ids 1

code/train.py

@@ -14,24 +14,25 @@ import random
 import os
 from concurrent import futures
 import threading
+import json
 
 
-def process_job(seed, args, experiment_id):
+def process_job(seed, parameters, experiment_id):
     logger = LoggerFactory.create(LOG_PATH, 'training_seed{}_ti{}'.format(
         seed, threading.get_ident()))
     logger.info('seed={}'.format(seed))
     seed_str = str(seed)
     experiment_id_str = str(experiment_id)
-    models_dir = args.models_dir + os.sep + experiment_id_str + os.sep + 'seeds' + \
+    models_dir = parameters['models_dir'] + os.sep + experiment_id_str + os.sep + 'seeds' + \
         os.sep + seed_str
     pathlib.Path(models_dir).mkdir(parents=True, exist_ok=True)
 
     dataset_parameters = DatasetParameters(
-        name=args.dataset_name,
-        test_size=args.test_size,
-        dev_size=args.dev_size,
+        name=parameters['dataset_name'],
+        test_size=parameters['test_size'],
+        dev_size=parameters['dev_size'],
         random_state=seed,
-        dataset_normalizer=args.dataset_normalizer
+        dataset_normalizer=parameters['dataset_normalizer']
     )
     dataset_parameters.save(models_dir, experiment_id_str)
 
@@ -39,17 +40,17 @@ def process_job(seed, args, experiment_id):
 
     trainer = Trainer(dataset)
 
-    for extracted_forest_size in args.extracted_forest_size:
+    for extracted_forest_size in parameters['extracted_forest_size']:
         logger.info('extracted_forest_size={}'.format(extracted_forest_size))
         sub_models_dir = models_dir + os.sep + 'extracted_forest_size' + os.sep + str(extracted_forest_size)
         pathlib.Path(sub_models_dir).mkdir(parents=True, exist_ok=True)
 
         model_parameters = ModelParameters(
-            forest_size=args.forest_size,
+            forest_size=parameters['forest_size'],
             extracted_forest_size=extracted_forest_size,
-            normalize_D=args.normalize_D,
-            subsets_used=args.subsets_used,
-            normalize_weights=args.normalize_weights,
+            normalize_D=parameters['normalize_D'],
+            subsets_used=parameters['subsets_used'],
+            normalize_weights=parameters['normalize_weights'],
             seed=seed
         )
         model_parameters.save(sub_models_dir, experiment_id)
@@ -63,6 +64,7 @@ if __name__ == "__main__":
     # get environment variables in .env
     load_dotenv(find_dotenv('.env.example'))
 
+    DEFAULT_EXPERIMENT_CONFIGURATION_PATH = 'experiments'
     DEFAULT_DATASET_NAME = 'boston'
     DEFAULT_NORMALIZE_D = False
     DEFAULT_DATASET_NORMALIZER = None
@@ -74,12 +76,14 @@ if __name__ == "__main__":
     DEFAULT_TEST_SIZE = 0.2
     DEFAULT_RANDOM_SEED_NUMBER = 1
     DEFAULT_SUBSETS_USED = 'train,dev'
-    DEFAULT_normalize_weights = False
+    DEFAULT_NORMALIZE_WEIGHTS = False
 
     begin_random_seed_range = 1
     end_random_seed_range = 2000
 
     parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    parser.add_argument('--experiment_configuration', nargs='?', type=str, default=None, help='Specify an experiment configuration file name. Overload all other parameters.')
+    parser.add_argument('--experiment_configuration_path', nargs='?', type=str, default=DEFAULT_EXPERIMENT_CONFIGURATION_PATH, help='Specify the experiment configuration directory path.')
     parser.add_argument('--dataset_name', nargs='?', type=str, default=DEFAULT_DATASET_NAME, help='Specify the dataset. Regression: boston, diabetes, linnerud, california_housing. Classification: iris, digits, wine, breast_cancer, olivetti_faces, 20newsgroups, 20newsgroups_vectorized, lfw_people, lfw_pairs, covtype, rcv1, kddcup99.')
     parser.add_argument('--normalize_D', action='store_true', default=DEFAULT_NORMALIZE_D, help='Specify if we want to normalize the prediction of the forest by doing the L2 division of the pred vectors.')
     parser.add_argument('--dataset_normalizer', nargs='?', type=str, default=DEFAULT_DATASET_NORMALIZER, help='Specify which dataset normalizer use (either standard, minmax, robust or normalizer).')
@@ -91,28 +95,50 @@ if __name__ == "__main__":
     parser.add_argument('--random_seed_number', nargs='?', type=int, default=DEFAULT_RANDOM_SEED_NUMBER, help='Number of random seeds used.')
     parser.add_argument('--seeds', nargs='+', type=int, default=None, help='Specific a list of seeds instead of generate them randomly')
     parser.add_argument('--subsets_used', nargs='+', type=str, default=DEFAULT_SUBSETS_USED, help='train,dev: forest on train, OMP on dev. train+dev,train+dev: both forest and OMP on train+dev. train,train+dev: forest on train+dev and OMP on dev.')
-    parser.add_argument('--normalize_weights', action='store_true', default=DEFAULT_normalize_weights, help='Divide the predictions by the weights sum.')
+    parser.add_argument('--normalize_weights', action='store_true', default=DEFAULT_NORMALIZE_WEIGHTS, help='Divide the predictions by the weights sum.')
     args = parser.parse_args()
 
-    pathlib.Path(args.models_dir).mkdir(parents=True, exist_ok=True)
+    if args.experiment_configuration:
+        with open(args.experiment_configuration_path + os.sep + \
+            args.experiment_configuration + '.json', 'r') as input_file:
+            parameters = json.load(input_file)
+    else:
+        parameters = args.__dict__
+
+    pathlib.Path(parameters['models_dir']).mkdir(parents=True, exist_ok=True)
 
     logger = LoggerFactory.create(LOG_PATH, os.path.basename(__file__))
 
-    args.extracted_forest_size = args.extracted_forest_size \
-        if type(args.extracted_forest_size) == list \
-        else [args.extracted_forest_size]
+    parameters['extracted_forest_size'] = parameters['extracted_forest_size'] \
+        if type(parameters['extracted_forest_size']) == list \
+        else [parameters['extracted_forest_size']]
 
-    if args.seeds != None and args.random_seed_number > 1:
+    if parameters['seeds'] != None and parameters['random_seed_number'] > 1:
         logger.warning('seeds and random_seed_number parameters are both specified. Seeds will be used.')    
 
-    seeds = args.seeds if args.seeds is not None \
+    seeds = parameters['seeds'] if parameters['seeds'] is not None \
         else [random.randint(begin_random_seed_range, end_random_seed_range) \
-        for i in range(args.random_seed_number)]
-
-    experiment_id = resolve_experiment_id(args.models_dir)
+        for i in range(parameters['random_seed_number'])]
 
+    # Resolve the next experiment id number (last id + 1)
+    experiment_id = resolve_experiment_id(parameters['models_dir'])
     logger.info('Experiment id: {}'.format(experiment_id))
 
+    """
+    If the experiment configuration isn't comming from
+    an already existing file, save it to a json file to
+    keep trace of it.
+    """
+    if args.experiment_configuration is None:
+        with open(args.experiment_configuration_path + os.sep + 'unnamed_{}.json'.format(
+            experiment_id), 'w') as output_file:
+            json.dump(
+                parameters,
+                output_file,
+                indent=4
+            )
+
+    # Train as much job as there are seeds
     with futures.ProcessPoolExecutor(len(seeds)) as executor:
         list(f.result() for f in futures.as_completed(executor.submit(process_job, seed,
-            args, experiment_id) for seed in seeds))
+            parameters, experiment_id) for seed in seeds))