1
|
{"args": {"mode": "bp", "dataset": "cifar10", "depth": 5, "width": 64, "act": "tanh", "residual": 1, "epochs": 5, "batch_size": 128, "train_examples": 0, "task_seed": 0, "task_train_examples": 50000, "task_test_examples": 10000, "task_levels": 8, "task_n_in": 128, "task_classes": 10, "teacher_depth": 8, "teacher_width": 64, "teacher_residual": 1, "eta": 0.05, "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, "predictor_mode": "diagonal", "normalize_delta": 0, "settle_steps": 0, "kappa": 0.0, "feedback": "error", "seed": 3, "device": "cuda", "log_every": 500, "max_steps": 0, "probe_bs": 256, "outdir": "results", "tag": "scale_v3_cifar10_bp_w64_d5_s3", "n_in": 3072, "n_out": 10}, "provenance": {"git_commit": "d4aef1a571c21e6e4e4125c90aa58da7f14e62e3", "git_dirty": false}, "steps": [{"step": 0, "epoch": 0, "train_loss": 2.5312681198120117}, {"epoch_end": 0, "step": 391, "test_acc": 0.3695, "test_loss": 1.789446826171875}, {"step": 500, "epoch": 1, "train_loss": 1.7008291482925415}, {"epoch_end": 1, "step": 782, "test_acc": 0.3846, "test_loss": 1.756138720703125}, {"step": 1000, "epoch": 2, "train_loss": 1.6484553813934326}, {"epoch_end": 2, "step": 1173, "test_acc": 0.3978, "test_loss": 1.7134104614257812}, {"step": 1500, "epoch": 3, "train_loss": 1.6287935972213745}, {"epoch_end": 3, "step": 1564, "test_acc": 0.403, "test_loss": 1.7025950805664063}, {"epoch_end": 4, "step": 1955, "test_acc": 0.4177, "test_loss": 1.6603008178710938}], "final": {"test_acc": 0.4177, "test_loss": 1.6603008178710938, "wall_s": 6.811361074447632}}
|
