|
pulsatrix 1.1.0
|
CEM's settings. The defaults are the paper's MNIST settings, scaled to its 1,000 steps. More...
#include <contrastive.hpp>
Public Attributes | |
| float | kappa = 0.0f |
| The margin κ the class change (or keep) must reach in the outputs. | |
| float | beta = 0.1f |
| The L1 weight β: larger gives sparser explanations. | |
| float | c_init = 10.0f |
| The first weight c of the prediction loss; the binary search moves it. | |
| int | search_steps = 9 |
| Rounds of the binary search over c. | |
| int | steps = 1000 |
| FISTA iterations in each round. | |
| float | learning_rate = 0.01f |
| std::vector< float > | baseline |
| Each feature's absence value; empty for 0. | |
| std::vector< float > | upper |
| The most a pertinent negative may raise each feature to; empty for no limit. | |
CEM's settings. The defaults are the paper's MNIST settings, scaled to its 1,000 steps.
| std::vector<float> pulsatrix::ContrastiveOptions::baseline |
Each feature's absence value; empty for 0.
| float pulsatrix::ContrastiveOptions::beta = 0.1f |
The L1 weight β: larger gives sparser explanations.
| float pulsatrix::ContrastiveOptions::c_init = 10.0f |
The first weight c of the prediction loss; the binary search moves it.
| float pulsatrix::ContrastiveOptions::kappa = 0.0f |
The margin κ the class change (or keep) must reach in the outputs.
| float pulsatrix::ContrastiveOptions::learning_rate = 0.01f |
| int pulsatrix::ContrastiveOptions::search_steps = 9 |
Rounds of the binary search over c.
| int pulsatrix::ContrastiveOptions::steps = 1000 |
FISTA iterations in each round.
| std::vector<float> pulsatrix::ContrastiveOptions::upper |
The most a pertinent negative may raise each feature to; empty for no limit.