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
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
|
#!/usr/bin/env bash
# Strict mechanical finalizer for all committed evidence. Future immutable
# gates are re-audited only after their gate files exist; this never opens an
# incomplete panel or generates a task endpoint.
set -eu
cd "$(dirname "$0")/.."
experiments/finalize_claims.sh
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/cifar_image_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/conv_local_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/bci_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/bci_td_protocol_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/bci_v2_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/bci_v2_recovery_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/bci_v2_calibrated_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/oral_a_dynamic_scaling_smoke.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 -m py_compile \
experiments/bci_td_run.py experiments/analyze_bci_td_development.py \
experiments/bci_td_confirmation.py \
experiments/analyze_bci_td_confirmation.py \
experiments/bci_v2_run.py experiments/analyze_bci_v2_development.py \
experiments/bci_v2_recovery_run.py \
experiments/analyze_bci_v2_recovery_development.py \
experiments/bci_v2_calibrated_run.py \
experiments/analyze_bci_v2_calibrated_development.py \
experiments/bci_v2_calibrated_confirmation.py \
experiments/analyze_bci_v2_calibrated_confirmation.py \
experiments/oral_a_dynamic_scaling.py \
experiments/analyze_oral_a_dynamic_scaling.py \
experiments/oral_a_dynamic_scaling_v2.py \
experiments/analyze_oral_a_dynamic_scaling_v2.py \
experiments/plot_resnet_confirmation.py \
experiments/plot_oral_a_scaling.py \
experiments/plot_bci_v2_confirmation.py \
experiments/audit_manuscript.py
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_kp_dynamic_projection.py >/dev/null
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_kp_dynamic_projection_short.py >/dev/null
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_kp_dynamic_projection_full.py >/dev/null
if [ -f results/kp_dynamic_projection_confirmation_gate.json ]; then
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_kp_dynamic_projection_confirmation.py >/dev/null
/home/yurenh2/miniconda3/bin/python \
experiments/plot_resnet_confirmation.py >/dev/null
jq -e '
.strict == true and
.gate_status == "passed" and
.required_seeds == [10, 11, 12, 13, 14] and
.statistics.dynamic_test_accuracy_mean >= 0.89 and
.statistics.clean_minus_dynamic_one_sided_95pct_upper_points <= 2.5 and
.statistics.dynamic_early_alignment_mean >= 0.90 and
.statistics.dynamic_mac_ratio_to_bp <= 1.34 and
.statistics.dynamic_logical_task_loss_queries == [0] and
.statistics.dynamic_neutral_observations_per_ordinary_example == [1] and
.statistics.dynamic_peak_allocated_gib_mean <= 2.5 and
.statistics.dynamic_wall_ratio_to_clean_kp_mean > 1
' results/figs/figure4_resnet_confirmation_manifest.json >/dev/null
fi
if [ -f results/bci_td_dev_gate.json ]; then
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_td_development.py >/dev/null
fi
if [ -f results/bci_td_confirmation_gate.json ]; then
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_td_confirmation.py >/dev/null
fi
if [ -f results/bci_v2_dev_gate.json ]; then
PYTHONPATH=. /home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_v2_development.py >/dev/null
fi
if [ -f results/bci_v2_recovery_dev_gate.json ]; then
PYTHONPATH=. /home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_v2_recovery_development.py >/dev/null
fi
if [ -f results/bci_v2_calibrated_dev_gate.json ]; then
PYTHONPATH=. /home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_v2_calibrated_development.py >/dev/null
fi
if [ -f results/bci_v2_calibrated_confirmation_gate.json ]; then
PYTHONPATH=. /home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_bci_v2_calibrated_confirmation.py >/dev/null
/home/yurenh2/miniconda3/bin/python \
experiments/plot_bci_v2_confirmation.py >/dev/null
jq -e '
.strict == true and
.gate_status == "passed" and
.task_seeds == [30, 31, 32, 33, 34, 35] and
.model_seeds == [0, 1, 2, 3, 4] and
.record_count == 30 and
.statistics.intact_final_mean >= 0.99 and
.statistics.residual_soma_corr_mean <= 0.10 and
.statistics.surrounding_accuracy_lower >= 0.50 and
.statistics.terminal_accuracy_lower >= 0.75 and
.statistics.outcome_lesion_drop_lower >= 0.15 and
.statistics.critic_expectedness_lower >= 0.02 and
.statistics.positive_sign_count == 30
' results/figs/figure5_bci_v2_manifest.json >/dev/null
fi
if [ -f results/oral_a_dynamic_scaling_gate.json ]; then
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_oral_a_dynamic_scaling.py >/dev/null
fi
if [ -f results/oral_a_dynamic_scaling_v2_gate.json ]; then
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/analyze_oral_a_dynamic_scaling_v2.py >/dev/null
/home/yurenh2/miniconda3/bin/python \
experiments/plot_oral_a_scaling.py >/dev/null
jq -e '
.strict == true and
.gate_status == "passed" and
.complete_grid == true and
.required_methods == ["bp", "dfa", "clean_kp", "dynamic"] and
.required_depths == [20, 32, 56] and
.required_seeds == [10, 11, 12, 13, 14] and
.record_count == 60 and
.statistics.dynamic_d20_to_d56_gain_points_mean >= 0.5 and
(.statistics.dynamic_d20_to_d56_gain_points | map(select(. > 0)) | length) == 5 and
(.statistics.dynamic_early_alignment_mean | [.[]] | min) >= 0.999 and
(.statistics.dynamic_mac_ratio_to_bp_mean | [.[]] | max) <= 1.34 and
.statistics.dynamic_logical_task_loss_queries == [0] and
.statistics.dynamic_neutral_observations_per_ordinary_example == [1]
' results/figs/figure6_standard_depth_scaling_manifest.json >/dev/null
fi
/home/yurenh2/miniconda3/bin/python \
experiments/plot_tracking_failure.py >/dev/null
/home/yurenh2/miniconda3/bin/python \
experiments/plot_dynamic_stability.py >/dev/null
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/audit_native_baselines.py \
--results results/native \
--output results/native_baseline_audit.md
/home/yurenh2/miniconda3/envs/ep_pascal/bin/python3 \
experiments/audit_manuscript.py >/dev/null
jq -e '
.strict == true and
.status == "passed" and
(.audited_numeric_claims | length) == 51 and
(.gates | map(select(.status == "passed")) | length) == 2 and
(.gates | map(select(.status == "failed")) | length) == 1 and
(.gates[] | select(.status == "failed") | .false_checks | length) == 7
' paper/manuscript_audit.json >/dev/null
git diff --check
echo "accept-bar mechanical audits passed"
|
