1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
|
{
"schema": "rain_ep_structured_bias_screen_v1",
"sdil": {
"revision": "65f617eaf07a87c84475ad0010dea844422163e1"
},
"author": {
"repository": "https://github.com/rain-neuromorphics/energy-based-learning",
"revision": "6b253fd8a5d267535f58ab79992256ef10031ceb"
},
"protocol": {
"dataset": "FashionMNIST",
"network": "author comparative-study ConvHopfieldEnergy32",
"network_protocol": "comparative32",
"algorithm": "equilibrium propagation",
"beta_policy": "fixed_positive",
"beta_seed": 7100,
"beta_value": 0.25,
"adapter": "dillavou",
"mode": "raw",
"bias_ratio": 1.0,
"dillavou_drift_ratio": 0.0,
"dillavou_calibration_steps": 1,
"predictor_rate": 1.0,
"neutral_cadence": 0,
"layer_calibration_steps": 1,
"layer_bias_normalization": "clean_difference",
"calibration_batches": 0,
"calibration_observations": 0,
"calibration_seconds": 7.152557373046875e-07,
"extra_equilibrium_phases_for_predictor": 0,
"predictor_neutral_source": "instruction_off_local_update_probe",
"bias_ratio_normalization": "initial_clean_local_update_rms_for_simulation_only",
"bias_ratio_normalization_visible_to_predictor": false,
"epochs": 10,
"train_limit": 60000,
"test_limit": 10000,
"evaluation_split": "test",
"data_seed": 6200,
"batch_size": 128,
"training_iterations": 15,
"inference_iterations": 60,
"schedule_epochs": 100,
"seed": 1988,
"device": "cuda",
"determinism": "best_effort_warn_only",
"autodiff_used_for_learning": false
},
"hardware": {
"torch_version": "2.3.1+cu118",
"torch_cuda_version": "11.8",
"cuda_visible_devices": "1",
"device_name": "NVIDIA GeForce GTX 1080"
},
"metrics": [
{
"epoch": 1,
"train_accuracy": 0.21611666666666668,
"train_cost": 2.952858947881063,
"test_accuracy": 0.1064,
"test_cost": 92706.8582,
"finite": true,
"corrector": {
"step": 468,
"bias_model": "dillavou_constant_update",
"clean_update_rms": 1.3665245006072042,
"bias_update_rms": 0.00015287711537102346,
"residual_update_rms": 0.000152877329837136,
"bias_to_clean_update_rms": 0.00011187294139482588,
"residual_to_clean_update_rms": 0.0001118730983375755,
"neutral_observations": 0
},
"wall_seconds": 832.7673797607422
},
{
"epoch": 2,
"train_accuracy": 0.09923333333333334,
"train_cost": NaN,
"test_accuracy": 0.1,
"test_cost": NaN,
"finite": false,
"corrector": {
"step": 937,
"bias_model": "dillavou_constant_update",
"clean_update_rms": NaN,
"bias_update_rms": NaN,
"residual_update_rms": NaN,
"bias_to_clean_update_rms": NaN,
"residual_to_clean_update_rms": NaN,
"neutral_observations": 0
},
"wall_seconds": 1665.599752664566
}
],
"epochs_completed": 2,
"beta_sign_counts": {
"positive": 938,
"negative": 0
},
"final": {
"epoch": 2,
"train_accuracy": 0.09923333333333334,
"train_cost": NaN,
"test_accuracy": 0.1,
"test_cost": NaN,
"finite": false,
"corrector": {
"step": 937,
"bias_model": "dillavou_constant_update",
"clean_update_rms": NaN,
"bias_update_rms": NaN,
"residual_update_rms": NaN,
"bias_to_clean_update_rms": NaN,
"residual_to_clean_update_rms": NaN,
"neutral_observations": 0
},
"wall_seconds": 1665.599752664566
}
}
|
