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
119
120
121
122
123
124
125
126
127
128
|
{
"schema": "rain_ep_structured_bias_screen_v1",
"sdil": {
"revision": "25f9b58f2fe478b7a4c728f404364ef8b8a92155"
},
"author": {
"repository": "https://github.com/rain-neuromorphics/energy-based-learning",
"revision": "6b253fd8a5d267535f58ab79992256ef10031ceb"
},
"protocol": {
"dataset": "FashionMNIST",
"network": "author ConvHopfieldEnergy28 32-64-10",
"algorithm": "equilibrium propagation",
"beta_policy": "fixed_positive",
"beta_seed": 7100,
"adapter": "layer",
"mode": "innovation",
"bias_ratio": 0.0002,
"predictor_rate": 0.2,
"neutral_cadence": 1,
"layer_calibration_steps": 1,
"layer_bias_normalization": "first_state",
"calibration_batches": 0,
"calibration_observations": 0,
"calibration_seconds": 2.384185791015625e-07,
"extra_equilibrium_phases_for_predictor": 0,
"predictor_neutral_source": "existing_first_EP_phase",
"bias_ratio_normalization": "initial_free_layer_state_rms",
"bias_ratio_normalization_visible_to_predictor": false,
"epochs": 3,
"train_limit": 10000,
"test_limit": 2000,
"evaluation_split": "train_holdout",
"data_seed": 6200,
"batch_size": 128,
"training_iterations": 12,
"inference_iterations": 30,
"seed": 1988,
"device": "cuda",
"determinism": "author_default",
"autodiff_used_for_learning": false
},
"hardware": {
"torch_version": "2.3.1+cu118",
"torch_cuda_version": "11.8",
"cuda_visible_devices": "7",
"device_name": "NVIDIA GeForce GTX 1080"
},
"metrics": [
{
"epoch": 1,
"train_accuracy": 0.5627,
"train_cost": 0.3277251761674881,
"test_accuracy": 0.7115,
"test_cost": 0.2376751184463501,
"finite": true,
"corrector": {
"step": 78,
"clean_state_difference_rms": 0.001959044826177622,
"bias_state_difference_rms": 5.9736316088780474e-05,
"residual_state_difference_rms": 6.459131498967555e-06,
"bias_to_clean_state_difference_rms": 0.030492572344725064,
"residual_to_clean_state_difference_rms": 0.0032970820333755447,
"neutral_observations": 128
},
"wall_seconds": 10.959503412246704
},
{
"epoch": 2,
"train_accuracy": 0.7526,
"train_cost": 0.209361084318161,
"test_accuracy": 0.757,
"test_cost": 0.20456177616119384,
"finite": true,
"corrector": {
"step": 157,
"clean_state_difference_rms": 0.0023768556991528105,
"bias_state_difference_rms": 6.0839851204849585e-05,
"residual_state_difference_rms": 6.838803277078925e-06,
"bias_to_clean_state_difference_rms": 0.025596779487511552,
"residual_to_clean_state_difference_rms": 0.0028772479875477926,
"neutral_observations": 128
},
"wall_seconds": 21.46092987060547
},
{
"epoch": 3,
"train_accuracy": 0.7888,
"train_cost": 0.1878269561290741,
"test_accuracy": 0.7775,
"test_cost": 0.1981061978340149,
"finite": true,
"corrector": {
"step": 236,
"clean_state_difference_rms": 0.0029137031845211728,
"bias_state_difference_rms": 6.27170698976096e-05,
"residual_state_difference_rms": 7.014578116605139e-06,
"bias_to_clean_state_difference_rms": 0.021524865755300427,
"residual_to_clean_state_difference_rms": 0.0024074442976448505,
"neutral_observations": 128
},
"wall_seconds": 32.3717303276062
}
],
"epochs_completed": 3,
"beta_sign_counts": {
"positive": 0,
"negative": 0
},
"final": {
"epoch": 3,
"train_accuracy": 0.7888,
"train_cost": 0.1878269561290741,
"test_accuracy": 0.7775,
"test_cost": 0.1981061978340149,
"finite": true,
"corrector": {
"step": 236,
"clean_state_difference_rms": 0.0029137031845211728,
"bias_state_difference_rms": 6.27170698976096e-05,
"residual_state_difference_rms": 7.014578116605139e-06,
"bias_to_clean_state_difference_rms": 0.021524865755300427,
"residual_to_clean_state_difference_rms": 0.0024074442976448505,
"neutral_observations": 128
},
"wall_seconds": 32.3717303276062
}
}
|
