|
pulsatrix 1.1.0
|
Whole-model LRP (LRP, LRPTarget, LRPSeed), rule selection (LRPRule, LRPRuleConfig), Zennit-style composites (lrp_composite), conservation checks, and BatchNormFold (folds an eval-mode BatchNorm into the preceding layer before LRP). More...

Files | |
| file | batch_norm_fold.hpp |
| Folds an eval-mode BatchNorm into the Conv2D before it for the duration of an LRP explanation – Zennit's BatchNorm canonizer (roadmap FND-5, lrp_issues #8). | |
| file | fine_tune_diff.hpp |
| What did a fine-tune change? The fine-tuned model's relevance minus the base model's for the same prediction, input by input and token by token across a dataset, and intruder dimensions: new top singular directions in the fine-tuned weights (TRN-11). | |
| file | lrp.hpp |
| LRP – whole-model Layer-wise Relevance Propagation explainer. | |
| file | lrp_conservation.hpp |
| Production API for LRP's conservation-delta audit metric (charter Phase 4). | |
| file | lrp_rule_config.hpp |
| Selects which LRP rule variant a Module::propagate_relevance() call uses. | |
Whole-model LRP (LRP, LRPTarget, LRPSeed), rule selection (LRPRule, LRPRuleConfig), Zennit-style composites (lrp_composite), conservation checks, and BatchNormFold (folds an eval-mode BatchNorm into the preceding layer before LRP).