The RESSPECT project is a result from an inter-collaboration agreement established between the Cosmostatistics Initiative (COIN) and the LSST Dark Energy Science Collaboration (DESC) with the goal of developing a recommendation system for telescope resource allocation able to optimize photometric supernova cosmology anaylsis.
Whenever trying to use QBDEntropy or QBDMI with batches in the data stored at
path_to_ini_files['train'] = '/media/RESSPECT/data/SNPCC/bazin/IAL2020/realistic/Train.csv' path_to_ini_files['test'] = '/media/RESSPECT/data/SNPCC/bazin/IAL2020/realistic/Test.csv' path_to_ini_files['validation'] = '/media/RESSPECT/data/SNPCC/bazin/IAL2020/realistic/Validation.csv' path_to_features_dir = '/media/RESSPECT/data/SNPCC/bazin/IAL2020/realistic/pool/'
I get the following error:
Processing night: 20
Before update_samples: ... train: 10 ... test: 1000 ... validation: 1000 ... pool: 3605
Inside classify: ... train_features: (10, 20) ... train_labels: (10,) ... pool_features: (3605, 20)
Metrics names: ['accuracy', 'efficiency', 'purity', 'fom'] Metrics values: [0.452, 0.7321428571428571, 0.25153374233128833, 0.07375394875394876] Traceback (most recent call last): File "learn_loop_timedomain.py", line 65, in
survey=survey, queryable=queryable, n_estimators=n_estimators)
File "/media/emille/git/COIN/RESSPECT_repo/RESSPECT/resspect/time_domain_loop.py", line 298, in time_domain_loop
screen=False)
File "/media/emille/git/COIN/RESSPECT_repo/RESSPECT/resspect/database.py", line 1219, in make_query_budget
criteria="entropy" )
File "/media/emille/git/COIN/RESSPECT_repo/RESSPECT/resspect/query_budget_strategies.py", line 163, in batch_queries_mi_entropy
probs_B_K_C = probs_B_K_C[pool_query_filter]
TypeError: 'NoneType' object is not subscriptable