Pre-Training Depth Feasibility Certificates / results.json

Failed on benchmark

Raw ⬇ ZIP
  1{
  2  "seed": 1319,
  3  "constants": {
  4    "floor": 0.08,
  5    "Csyn": 0.8,
  6    "Cmeta": 0.12,
  7    "m": 2.0,
  8    "epsilon": 0.16
  9  },
 10  "scaling_checks": {
 11    "depth_inverse_prediction": {
 12      "predicted_log_slope": -1.0,
 13      "observed_log_slope": -0.998895762481767,
 14      "scaled_depth_min": 0.7995687333596901,
 15      "scaled_depth_max": 0.8100294955481461,
 16      "relative_spread": 0.013042510938075805
 17    },
 18    "metadata_exponential_prediction": {
 19      "predicted_ratio_for_delta_s_2": 0.5,
 20      "observed_ratios": [
 21        0.49879613159473835,
 22        0.5016210175932083,
 23        0.4950466275605574,
 24        0.47835715693957853
 25      ],
 26      "max_abs_ratio_error": 0.02164284306042147
 27    },
 28    "threshold_prediction": {
 29      "predicted_continuous_boundary_D": 11.510791366906474,
 30      "observed_first_grid_passing_D": 16,
 31      "epsilon": 0.16
 32    }
 33  },
 34  "calibration": {
 35    "floor_bracket_L_U": [
 36      0.07597848545363584,
 37      0.08397848545363584
 38    ],
 39    "fit_coefficients": [
 40      0.08297848545363584,
 41      0.7989193055645585,
 42      0.12015191422269511
 43    ],
 44    "residual_std": 0.0014804082298792022
 45  },
 46  "screening": {
 47    "total_candidates": 90,
 48    "doomed": 66,
 49    "successful": 24,
 50    "rejected_doomed": 0,
 51    "rejection_rate_among_doomed": 0.0,
 52    "retained_success_rate": 1.0,
 53    "certificate_evaluations": 68,
 54    "evaluations_avoided_vs_exhaustive": 22,
 55    "feasible_certificates": 22,
 56    "feasible_certificate_precision": 1.0,
 57    "false_rejections": 0
 58  },
 59  "sample_records": [
 60    {
 61      "D": 2,
 62      "s": 0,
 63      "bits": 2,
 64      "upper": 0.6207784854536359,
 65      "lower": -0.4608215145463642,
 66      "actual": 0.6168,
 67      "success": false,
 68      "decision": "unresolved"
 69    },
 70    {
 71      "D": 2,
 72      "s": 0,
 73      "bits": 4,
 74      "upper": 0.6115784854536359,
 75      "lower": -0.45162151454636423,
 76      "actual": 0.6076,
 77      "success": false,
 78      "decision": "unresolved"
 79    },
 80    {
 81      "D": 2,
 82      "s": 0,
 83      "bits": 8,
 84      "upper": 0.6069784854536359,
 85      "lower": -0.4470215145463642,
 86      "actual": 0.603,
 87      "success": false,
 88      "decision": "unresolved"
 89    },
 90    {
 91      "D": 2,
 92      "s": 2,
 93      "bits": 2,
 94      "upper": 0.5607784854536358,
 95      "lower": -0.40082151454636417,
 96      "actual": 0.5568,
 97      "success": false,
 98      "decision": "unresolved"
 99    },
100    {
101      "D": 2,
102      "s": 2,
103      "bits": 4,
104      "upper": 0.5515784854536359,
105      "lower": -0.3916215145463642,
106      "actual": 0.5476,
107      "success": false,
108      "decision": "unresolved"
109    },
110    {
111      "D": 2,
112      "s": 2,
113      "bits": 8,
114      "upper": 0.5469784854536359,
115      "lower": -0.3870215145463642,
116      "actual": 0.543,
117      "success": false,
118      "decision": "unresolved"
119    },
120    {
121      "D": 2,
122      "s": 4,
123      "bits": 2,
124      "upper": 0.5307784854536359,
125      "lower": -0.3708215145463642,
126      "actual": 0.5268,
127      "success": false,
128      "decision": "unresolved"
129    },
130    {
131      "D": 2,
132      "s": 4,
133      "bits": 4,
134      "upper": 0.5215784854536358,
135      "lower": -0.3616215145463642,
136      "actual": 0.5176000000000001,
137      "success": false,
138      "decision": "unresolved"
139    }
140  ]
141}