{ "fold_results": [ { "fold": 0, "best_params": { "dropout": 0.19323529834582587, "lr": 0.000980484886752915, "weight_decay": 0.00014367509113739864, "backbone_lr_ratio": 0.13093310334961422, "moe_n_experts": 2, "moe_top_k": 2, "moe_expert_hidden_mult": 2, "llm_lora_r": 8, "d_model": 256, "num_heads": 8, "n_attn_layers": 4, "fusion_strategy": "attention", "head_hidden_dim": 128, "set_transformer_block": "sab" }, "epoch_mean": 11, "test_metrics": { "size": { "n_samples": 83, "mse": 1.5968842484584136, "rmse": 1.263678854954222, "mae": 0.5682635198246284, "r2": 0.022667532793975265 }, "delivery": { "n_samples": 58, "mse": 0.5483822645024997, "rmse": 0.740528368465719, "mae": 0.5250802371843621, "r2": 0.3029730502833111 }, "pdi": { "n_samples": 84, "accuracy": 0.7142857142857143, "precision": 0.6737967914438503, "recall": 0.7084818246614397, "f1": 0.6785714285714286 }, "ee": { "n_samples": 84, "accuracy": 0.7380952380952381, "precision": 0.7073879551820729, "recall": 0.7647619047619049, "f1": 0.7124896608767578 }, "toxic": { "n_samples": 58, "accuracy": 0.9827586206896551, "precision": 0.9912280701754386, "recall": 0.75, "f1": 0.8289085545722714 }, "biodist": { "n_samples": 58, "kl_divergence": 0.1971223395298808, "js_divergence": 0.043043902163233595 } } }, { "fold": 1, "best_params": { "dropout": 0.19323529834582587, "lr": 0.000980484886752915, "weight_decay": 0.00014367509113739864, "backbone_lr_ratio": 0.13093310334961422, "moe_n_experts": 2, "moe_top_k": 2, "moe_expert_hidden_mult": 2, "llm_lora_r": 8, "d_model": 256, "num_heads": 8, "n_attn_layers": 4, "fusion_strategy": "attention", "head_hidden_dim": 128, "set_transformer_block": "sab" }, "epoch_mean": 11, "test_metrics": { "size": { "n_samples": 84, "mse": 0.30847479651898757, "rmse": 0.5554050742647096, "mae": 0.41110294015102444, "r2": 0.05802764121606785 }, "delivery": { "n_samples": 61, "mse": 0.9177343265746907, "rmse": 0.9579845127008529, "mae": 0.589223776898179, "r2": 0.30534127327020133 }, "pdi": { "n_samples": 84, "accuracy": 0.6785714285714286, "precision": 0.5875, "recall": 0.5952380952380952, "f1": 0.5902439024390244 }, "ee": { "n_samples": 84, "accuracy": 0.5833333333333334, "precision": 0.5230357877416701, "recall": 0.5150801579373008, "f1": 0.49626374349858454 }, "toxic": { "n_samples": 61, "accuracy": 0.9836065573770492, "precision": 0.9915254237288136, "recall": 0.8333333333333333, "f1": 0.8957264957264958 }, "biodist": { "n_samples": 61, "kl_divergence": 0.13675394519783235, "js_divergence": 0.03248669854336822 } } }, { "fold": 2, "best_params": { "dropout": 0.19323529834582587, "lr": 0.000980484886752915, "weight_decay": 0.00014367509113739864, "backbone_lr_ratio": 0.13093310334961422, "moe_n_experts": 2, "moe_top_k": 2, "moe_expert_hidden_mult": 2, "llm_lora_r": 8, "d_model": 256, "num_heads": 8, "n_attn_layers": 4, "fusion_strategy": "attention", "head_hidden_dim": 128, "set_transformer_block": "sab" }, "epoch_mean": 11, "test_metrics": { "size": { "n_samples": 84, "mse": 0.3450456742192763, "rmse": 0.5874058854142307, "mae": 0.46413040039173903, "r2": 0.31854095833662033 }, "delivery": { "n_samples": 60, "mse": 0.633114832363136, "rmse": 0.7956851339337289, "mae": 0.5481160653211797, "r2": 0.15721442419210252 }, "pdi": { "n_samples": 84, "accuracy": 0.7142857142857143, "precision": 0.6517857142857143, "recall": 0.6744868035190617, "f1": 0.6583050847457628 }, "ee": { "n_samples": 84, "accuracy": 0.6428571428571429, "precision": 0.5853276353276353, "recall": 0.6349206349206349, "f1": 0.5970747928194737 }, "toxic": { "n_samples": 61, "accuracy": 1.0, "precision": 1.0, "recall": 1.0, "f1": 1.0 }, "biodist": { "n_samples": 60, "kl_divergence": 0.1542080131383728, "js_divergence": 0.03808539778255157 } } }, { "fold": 3, "best_params": { "dropout": 0.19323529834582587, "lr": 0.000980484886752915, "weight_decay": 0.00014367509113739864, "backbone_lr_ratio": 0.13093310334961422, "moe_n_experts": 2, "moe_top_k": 2, "moe_expert_hidden_mult": 2, "llm_lora_r": 8, "d_model": 256, "num_heads": 8, "n_attn_layers": 4, "fusion_strategy": "attention", "head_hidden_dim": 128, "set_transformer_block": "sab" }, "epoch_mean": 11, "test_metrics": { "size": { "n_samples": 83, "mse": 1.763092308495774, "rmse": 1.3278148622815509, "mae": 0.5504888193233006, "r2": 0.0027815521929155373 }, "delivery": { "n_samples": 59, "mse": 0.6925828062408207, "rmse": 0.8322156008155703, "mae": 0.5598128846357182, "r2": 0.32672722594328685 }, "pdi": { "n_samples": 84, "accuracy": 0.7380952380952381, "precision": 0.6612903225806452, "recall": 0.6612903225806452, "f1": 0.6612903225806452 }, "ee": { "n_samples": 84, "accuracy": 0.5952380952380952, "precision": 0.5301311591634171, "recall": 0.5438914027149321, "f1": 0.5247639796659405 }, "toxic": { "n_samples": 60, "accuracy": 0.9333333333333333, "precision": 0.7142857142857143, "recall": 0.9649122807017544, "f1": 0.7818181818181817 }, "biodist": { "n_samples": 60, "kl_divergence": 0.15787098167920866, "js_divergence": 0.04007834983720861 } } }, { "fold": 4, "best_params": { "dropout": 0.19323529834582587, "lr": 0.000980484886752915, "weight_decay": 0.00014367509113739864, "backbone_lr_ratio": 0.13093310334961422, "moe_n_experts": 2, "moe_top_k": 2, "moe_expert_hidden_mult": 2, "llm_lora_r": 8, "d_model": 256, "num_heads": 8, "n_attn_layers": 4, "fusion_strategy": "attention", "head_hidden_dim": 128, "set_transformer_block": "sab" }, "epoch_mean": 11, "test_metrics": { "size": { "n_samples": 84, "mse": 0.5697698999186622, "rmse": 0.75483104064331, "mae": 0.45241403512634515, "r2": 0.23757709942203908 }, "delivery": { "n_samples": 58, "mse": 0.8494501546811344, "rmse": 0.9216562019978677, "mae": 0.6002610280852894, "r2": 0.17859572323309747 }, "pdi": { "n_samples": 84, "accuracy": 0.7380952380952381, "precision": 0.6724137931034483, "recall": 0.6906158357771262, "f1": 0.6791666666666667 }, "ee": { "n_samples": 84, "accuracy": 0.6785714285714286, "precision": 0.6324786324786325, "recall": 0.6799145299145298, "f1": 0.6406077872744539 }, "toxic": { "n_samples": 59, "accuracy": 1.0, "precision": 1.0, "recall": 1.0, "f1": 1.0 }, "biodist": { "n_samples": 58, "kl_divergence": 0.17354171533022728, "js_divergence": 0.046613494801719065 } } } ], "summary_stats": { "size": { "mse_mean": 0.9166533855222226, "mse_std": 0.6318413727420791, "rmse_mean": 0.8978271435116048, "rmse_std": 0.33250534716905716, "mae_mean": 0.4892799429634075, "mae_std": 0.06014677111861511, "r2_mean": 0.12791895679232362, "r2_std": 0.1264782068138924 }, "delivery": { "mse_mean": 0.7282528768724562, "mse_std": 0.13657894316164607, "rmse_mean": 0.8496139635827478, "rmse_std": 0.08005616626887421, "mae_mean": 0.5644987984249458, "mae_std": 0.027329238310088152, "r2_mean": 0.2541703393843998, "r2_std": 0.07124134045642266 }, "pdi": { "accuracy_mean": 0.7166666666666667, "accuracy_std": 0.021821789023599242, "precision_mean": 0.6493573242827317, "precision_std": 0.03194842845400596, "recall_mean": 0.6660225763552736, "recall_std": 0.038761053417086286, "f1_mean": 0.6535154810007054, "f1_std": 0.03277945989198496 }, "ee": { "accuracy_mean": 0.6476190476190476, "accuracy_std": 0.05664465358747772, "precision_mean": 0.5956722339786855, "precision_std": 0.06857668680486774, "recall_mean": 0.6277137260498604, "recall_std": 0.09085325454039572, "f1_mean": 0.5942399928270421, "f1_std": 0.07818855962865172 }, "toxic": { "accuracy_mean": 0.9799397022800076, "accuracy_std": 0.02448827002026046, "precision_mean": 0.9394078416379934, "precision_std": 0.11262714591271072, "recall_mean": 0.9096491228070175, "recall_std": 0.10069013931425785, "f1_mean": 0.9012906464233896, "f1_std": 0.0883525548692155 }, "biodist": { "kl_divergence_mean": 0.16389939897510436, "kl_divergence_std": 0.020316882039832528, "js_divergence_mean": 0.0400615686256162, "js_divergence_std": 0.0047561799696747425 } } }