mirror of
https://github.com/mcloh/decision-models-benchmark.git
synced 2026-11-06 06:14:11 +00:00
328ccd1c2e
- harness/evaluate.py: partição congelada com SHA-256, amostra com seed, lote 1, truncamento comum, contagem de tokens, latência (sem aquecimento), pico de RAM e GPU; teste bloqueado até o congelamento da configuração. - analysis/metrics.py: acurácia, macro F1, Brier, NLL, ECE, acurácia seletiva. - Diagnósticos por resposta (metadata.diagnostics); Laya informa opções colapsadas pelo orçamento de cabeçalho; head_max_len via DMB_LAYA_HEAD_MAX_LEN. - infra/ds/collect.py e gpu_queue --env para pilotos e rodadas. - .gitignore: de models/ só metadados (cards, configs, licenças, índices); pesos, tokenizadores grandes, runtimes, fontes clonados e caches ignorados. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
100 lines
3.1 KiB
JSON
100 lines
3.1 KiB
JSON
{
|
|
"architecture": "boundary",
|
|
"architecture_version": 1,
|
|
"architectures": [
|
|
"BoundaryExtractor"
|
|
],
|
|
"attn_implementation": "sdpa",
|
|
"boundary_head": {
|
|
"abstention_loss_weight": 0.2,
|
|
"abstention_threshold": 0.5,
|
|
"adaptive_threshold": false,
|
|
"bidirectional_proposals": true,
|
|
"boundary_attention_heads": 4,
|
|
"boundary_attention_layers": 2,
|
|
"boundary_attention_window": 128,
|
|
"boundary_dim": 128,
|
|
"boundary_ffn_multiplier": 2.0,
|
|
"boundary_focal_clip": 0.05,
|
|
"boundary_focal_gamma_negative": 2.0,
|
|
"boundary_focal_gamma_positive": 0.0,
|
|
"boundary_marginal_loss": "asymmetric_focal",
|
|
"boundary_negative_weight": 0.5,
|
|
"boundary_refinement_layers": 1,
|
|
"boundary_top_k_alpha": 0.08,
|
|
"boundary_top_k_bucket": 8,
|
|
"boundary_top_k_max": 128,
|
|
"candidate_attention_heads": 4,
|
|
"candidate_attention_layers": 0,
|
|
"candidate_budget": 192,
|
|
"candidate_pool": "shared",
|
|
"classification_loss_weight": 1.0,
|
|
"classification_temperature": 1.0,
|
|
"consistency_loss_weight": 0.1,
|
|
"consistency_warmup_steps": 2000,
|
|
"content_dim": 64,
|
|
"content_soft_max_pool": false,
|
|
"count_loss_weight": 0.2,
|
|
"directional_relation_states": true,
|
|
"dropout": 0.1,
|
|
"enable_abstention": true,
|
|
"enable_count_head": true,
|
|
"enable_records": true,
|
|
"enable_relations": true,
|
|
"enable_rotary_endpoints": true,
|
|
"enable_span_content": true,
|
|
"end_block_size": 256,
|
|
"end_top_k": 24,
|
|
"endpoint_difference_features": true,
|
|
"ends_per_start": 12,
|
|
"export_mode": "auto",
|
|
"hard_negative_keep_all_when_absent": true,
|
|
"hard_negatives_per_positive": 20,
|
|
"loss_reduction": "sum",
|
|
"max_gold_per_query": 64,
|
|
"max_negative_queries_per_batch": 64,
|
|
"min_pool_per_query": 8,
|
|
"minimum_hard_negatives": 16,
|
|
"multihead_pair_compat_heads": 8,
|
|
"negative_query_ratio": 1.0,
|
|
"overlap_policy": "flat",
|
|
"pair_dim": 128,
|
|
"pair_temperature": 1.0,
|
|
"pool_boundary_top_k": 32,
|
|
"pool_size": 192,
|
|
"proposal_loss_weight": 0.3,
|
|
"query_attention_layers": 0,
|
|
"query_conditioned_inside_weight": true,
|
|
"record_anchor_proposal_threshold": 0.2,
|
|
"record_anchor_threshold": 0.5,
|
|
"record_dim": 128,
|
|
"record_field_threshold": 0.5,
|
|
"record_instance_queries": 32,
|
|
"record_loss_weight": 1.0,
|
|
"record_temperature": 1.0,
|
|
"relation_argument_proposal_threshold": 0.2,
|
|
"relation_biaffine_content": true,
|
|
"relation_heads_per_type": 32,
|
|
"relation_loss_weight": 1.0,
|
|
"relation_pair_cap": 64,
|
|
"relation_tails_per_type": 32,
|
|
"relation_temperature": 1.0,
|
|
"rerank_listwise_weight": 0.3,
|
|
"reranker_endpoint_compat": true,
|
|
"rotary_base": 10000.0,
|
|
"soft_iou_anneal_steps": 20000,
|
|
"soft_iou_aux_weight": 0.2,
|
|
"start_top_k": 24,
|
|
"starts_per_end": 12,
|
|
"training_candidate_budget": 192,
|
|
"use_inside_evidence": true,
|
|
"vectorized_pair_elements": 16777216
|
|
},
|
|
"config_version": 3,
|
|
"max_len": 4096,
|
|
"model_name": "microsoft/mdeberta-v3-base",
|
|
"model_type": "extractor",
|
|
"token_pooling": "first",
|
|
"transformers_version": "5.17.0"
|
|
}
|