-
Notifications
You must be signed in to change notification settings - Fork 43
Open
Description
The only time in which positive_example_penalty appears in the codebase is in:
plasma-python/plasma/conf_parser.py
Lines 86 to 102 in c82ba61
| # ensure shallow model has +1 -1 target. | |
| if params['model']['shallow'] or params['target'] == 'hinge': | |
| params['data']['target'] = HingeTarget | |
| elif params['target'] == 'maxhinge': | |
| MaxHingeTarget.fac = params['data']['positive_example_penalty'] | |
| params['data']['target'] = MaxHingeTarget | |
| elif params['target'] == 'binary': | |
| params['data']['target'] = BinaryTarget | |
| elif params['target'] == 'ttd': | |
| params['data']['target'] = TTDTarget | |
| elif params['target'] == 'ttdinv': | |
| params['data']['target'] = TTDInvTarget | |
| elif params['target'] == 'ttdlinear': | |
| params['data']['target'] = TTDLinearTarget | |
| else: | |
| g.print_unique('Unkown type of target. Exiting') | |
| exit(1) |
which is loaded only for an unused method in the
MaxHingeTarget class, noted here:plasma-python/plasma/models/targets.py
Line 153 in 7986f46
| def loss_np(y_true, y_pred): |
Need to extend it to the other target functions, remove it as a parameter, or document this more thoroughly.
Not as important for DIII-D datasets as it is for our JET datasets (for which the non-/disruptive classes are more imbalanced).
- Also,
conf['training']['ranking_difficulty_fac']: 1.0 # how much to upweight incorrectly classified shots during trainingappears to perform a very related role, but instead withinloader.py,mpi_runner.py, andperformance.py.
Metadata
Metadata
Assignees
Labels
No labels