1 files changed, 1 insertions, 0 deletions
diff --git a/results/depthtask_dev_v1_hierarchical_bp_w16_d12_t0_s0.json b/results/depthtask_dev_v1_hierarchical_bp_w16_d12_t0_s0.json
new file mode 100644
index 0000000..b497760
--- /dev/null
+++ b/results/depthtask_dev_v1_hierarchical_bp_w16_d12_t0_s0.json
@@ -0,0 +1 @@
+{"args": {"mode": "bp", "dataset": "hierarchical", "depth": 12, "width": 16, "act": "tanh", "residual": 1, "epochs": 25, "batch_size": 128, "train_examples": 0, "val_examples": 5000, "split_seed": 2027, "eval_split": "validation", "task_seed": 0, "task_train_examples": 50000, "task_test_examples": 10000, "task_levels": 6, "task_n_in": 128, "task_classes": 10, "teacher_depth": 8, "teacher_width": 64, "teacher_residual": 1, "eta": 0.03, "eta_A": 0.02, "eta_P": 0.002, "momentum": 0.9, "w_scale": 1.0, "a_scale": 1.0, "feedback_scale": 1.0, "pert_sigma": 0.01, "pert_every": 4, "pert_ndirs": 16, "pert_mode": "simultaneous", "use_residual": 1, "raw_scale_control": "none", "learn_A": 1, "learn_P": 1, "p_neutral": 1, "p_warmup_steps": 200, "p_warmup_eta": 0.05, "nuis_rho": 0.0, "traffic_mode": "soma", "predictor_mode": "diagonal", "normalize_delta": 0, "settle_steps": 0, "kappa": 0.0, "feedback": "error", "seed": 0, "device": "cuda", "log_every": 200, "max_steps": 0, "probe_bs": 512, "outdir": "results", "tag": "depthtask_dev_v1_hierarchical_bp_w16_d12_t0_s0", "n_in": 64, "n_out": 10}, "split": {"dataset": "hierarchical", "task_seed": 0, "train_examples": 45000, "test_examples": 10000, "evaluation_split": "validation", "synthetic_generator": true, "split_seed": 2027, "validation_examples": 5000, "validation_index_sha256": "f5282ed201086357dbe24c23e70cc9eb86deaaf70d99c41553ffc78626ee8d6c", "validation_class_counts": {"0": 500, "1": 500, "2": 500, "3": 500, "4": 500, "5": 500, "6": 500, "7": 500, "8": 500, "9": 500}, "split_from_training_only": true}, "provenance": {"git_commit": "a0696a3d72cc5bf69876e1716b74f7e1ac525dd0", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.582568407058716}, {"step": 200, "epoch": 0, "train_loss": 1.779198169708252}, {"epoch_end": 0, "step": 352, "eval_split": "validation", "eval_acc": 0.326, "eval_loss": 1.924127734375, "val_acc": 0.326, "val_loss": 1.924127734375}, {"step": 400, "epoch": 1, "train_loss": 1.7927218675613403}, {"step": 600, "epoch": 1, "train_loss": 1.7545812129974365}, {"epoch_end": 1, "step": 704, "eval_split": "validation", "eval_acc": 0.3358, "eval_loss": 1.9051428466796876, "val_acc": 0.3358, "val_loss": 1.9051428466796876}, {"step": 800, "epoch": 2, "train_loss": 1.7563821077346802}, {"step": 1000, "epoch": 2, "train_loss": 1.849263072013855}, {"epoch_end": 2, "step": 1056, "eval_split": "validation", "eval_acc": 0.337, "eval_loss": 1.8839279541015626, "val_acc": 0.337, "val_loss": 1.8839279541015626}, {"step": 1200, "epoch": 3, "train_loss": 1.6742051839828491}, {"step": 1400, "epoch": 3, "train_loss": 1.6528172492980957}, {"epoch_end": 3, "step": 1408, "eval_split": "validation", "eval_acc": 0.3476, "eval_loss": 1.8585701904296874, "val_acc": 0.3476, "val_loss": 1.8585701904296874}, {"step": 1600, "epoch": 4, "train_loss": 1.6684035062789917}, {"epoch_end": 4, "step": 1760, "eval_split": "validation", "eval_acc": 0.3682, "eval_loss": 1.8213005126953126, "val_acc": 0.3682, "val_loss": 1.8213005126953126}, {"step": 1800, "epoch": 5, "train_loss": 1.918573260307312}, {"step": 2000, "epoch": 5, "train_loss": 1.6686073541641235}, {"epoch_end": 5, "step": 2112, "eval_split": "validation", "eval_acc": 0.3652, "eval_loss": 1.8040160400390626, "val_acc": 0.3652, "val_loss": 1.8040160400390626}, {"step": 2200, "epoch": 6, "train_loss": 1.5829106569290161}, {"step": 2400, "epoch": 6, "train_loss": 1.672957420349121}, {"epoch_end": 6, "step": 2464, "eval_split": "validation", "eval_acc": 0.3664, "eval_loss": 1.800686328125, "val_acc": 0.3664, "val_loss": 1.800686328125}, {"step": 2600, "epoch": 7, "train_loss": 1.6874823570251465}, {"step": 2800, "epoch": 7, "train_loss": 1.7868740558624268}, {"epoch_end": 7, "step": 2816, "eval_split": "validation", "eval_acc": 0.3718, "eval_loss": 1.78406845703125, "val_acc": 0.3718, "val_loss": 1.78406845703125}, {"step": 3000, "epoch": 8, "train_loss": 1.629400610923767}, {"epoch_end": 8, "step": 3168, "eval_split": "validation", "eval_acc": 0.3708, "eval_loss": 1.779471875, "val_acc": 0.3708, "val_loss": 1.779471875}, {"step": 3200, "epoch": 9, "train_loss": 1.6815563440322876}, {"step": 3400, "epoch": 9, "train_loss": 1.5809072256088257}, {"epoch_end": 9, "step": 3520, "eval_split": "validation", "eval_acc": 0.3728, "eval_loss": 1.7811262451171874, "val_acc": 0.3728, "val_loss": 1.7811262451171874}, {"step": 3600, "epoch": 10, "train_loss": 1.6039706468582153}, {"step": 3800, "epoch": 10, "train_loss": 1.4702858924865723}, {"epoch_end": 10, "step": 3872, "eval_split": "validation", "eval_acc": 0.377, "eval_loss": 1.777668994140625, "val_acc": 0.377, "val_loss": 1.777668994140625}, {"step": 4000, "epoch": 11, "train_loss": 1.631117343902588}, {"step": 4200, "epoch": 11, "train_loss": 1.6752244234085083}, {"epoch_end": 11, "step": 4224, "eval_split": "validation", "eval_acc": 0.3782, "eval_loss": 1.7660619140625, "val_acc": 0.3782, "val_loss": 1.7660619140625}, {"step": 4400, "epoch": 12, "train_loss": 1.7986152172088623}, {"epoch_end": 12, "step": 4576, "eval_split": "validation", "eval_acc": 0.3752, "eval_loss": 1.784043115234375, "val_acc": 0.3752, "val_loss": 1.784043115234375}, {"step": 4600, "epoch": 13, "train_loss": 1.529523253440857}, {"step": 4800, "epoch": 13, "train_loss": 1.677403211593628}, {"epoch_end": 13, "step": 4928, "eval_split": "validation", "eval_acc": 0.3686, "eval_loss": 1.7831521240234376, "val_acc": 0.3686, "val_loss": 1.7831521240234376}, {"step": 5000, "epoch": 14, "train_loss": 1.3653477430343628}, {"step": 5200, "epoch": 14, "train_loss": 1.5780153274536133}, {"epoch_end": 14, "step": 5280, "eval_split": "validation", "eval_acc": 0.3754, "eval_loss": 1.7836129150390625, "val_acc": 0.3754, "val_loss": 1.7836129150390625}, {"step": 5400, "epoch": 15, "train_loss": 1.7597254514694214}, {"step": 5600, "epoch": 15, "train_loss": 1.5191596746444702}, {"epoch_end": 15, "step": 5632, "eval_split": "validation", "eval_acc": 0.3788, "eval_loss": 1.7681939453125, "val_acc": 0.3788, "val_loss": 1.7681939453125}, {"step": 5800, "epoch": 16, "train_loss": 1.517289638519287}, {"epoch_end": 16, "step": 5984, "eval_split": "validation", "eval_acc": 0.3768, "eval_loss": 1.7822583251953126, "val_acc": 0.3768, "val_loss": 1.7822583251953126}, {"step": 6000, "epoch": 17, "train_loss": 1.5022385120391846}, {"step": 6200, "epoch": 17, "train_loss": 1.4912019968032837}, {"epoch_end": 17, "step": 6336, "eval_split": "validation", "eval_acc": 0.3724, "eval_loss": 1.7875509521484374, "val_acc": 0.3724, "val_loss": 1.7875509521484374}, {"step": 6400, "epoch": 18, "train_loss": 1.4263951778411865}, {"step": 6600, "epoch": 18, "train_loss": 1.5325927734375}, {"epoch_end": 18, "step": 6688, "eval_split": "validation", "eval_acc": 0.3876, "eval_loss": 1.772078955078125, "val_acc": 0.3876, "val_loss": 1.772078955078125}, {"step": 6800, "epoch": 19, "train_loss": 1.579502820968628}, {"step": 7000, "epoch": 19, "train_loss": 1.3733022212982178}, {"epoch_end": 19, "step": 7040, "eval_split": "validation", "eval_acc": 0.3842, "eval_loss": 1.756608349609375, "val_acc": 0.3842, "val_loss": 1.756608349609375}, {"step": 7200, "epoch": 20, "train_loss": 1.6390156745910645}, {"epoch_end": 20, "step": 7392, "eval_split": "validation", "eval_acc": 0.3814, "eval_loss": 1.7701624755859375, "val_acc": 0.3814, "val_loss": 1.7701624755859375}, {"step": 7400, "epoch": 21, "train_loss": 1.5420736074447632}, {"step": 7600, "epoch": 21, "train_loss": 1.5902906656265259}, {"epoch_end": 21, "step": 7744, "eval_split": "validation", "eval_acc": 0.3864, "eval_loss": 1.770354248046875, "val_acc": 0.3864, "val_loss": 1.770354248046875}, {"step": 7800, "epoch": 22, "train_loss": 1.6107194423675537}, {"step": 8000, "epoch": 22, "train_loss": 1.5532841682434082}, {"epoch_end": 22, "step": 8096, "eval_split": "validation", "eval_acc": 0.3852, "eval_loss": 1.7620365478515625, "val_acc": 0.3852, "val_loss": 1.7620365478515625}, {"step": 8200, "epoch": 23, "train_loss": 1.5031737089157104}, {"step": 8400, "epoch": 23, "train_loss": 1.6291133165359497}, {"epoch_end": 23, "step": 8448, "eval_split": "validation", "eval_acc": 0.3832, "eval_loss": 1.7563646728515625, "val_acc": 0.3832, "val_loss": 1.7563646728515625}, {"step": 8600, "epoch": 24, "train_loss": 1.5404670238494873}, {"epoch_end": 24, "step": 8800, "eval_split": "validation", "eval_acc": 0.3896, "eval_loss": 1.7488685791015626, "val_acc": 0.3896, "val_loss": 1.7488685791015626}], "final": {"eval_split": "validation", "eval_acc": 0.3896, "eval_loss": 1.7488685791015626, "wall_s": 59.528953313827515, "val_acc": 0.3896, "val_loss": 1.7488685791015626}}
+\ No newline at end of file