Merge nucleic/lucid-north-quail-rnvt into dev

2026-07-30 03:48:00 -07:00
parent e2a5314f1b
commit 9c43be6df1
17 changed files with 3325 additions and 13 deletions
@@ -0,0 +1,147 @@
+{
+  "curation": {
+    "excludedDuplicates": {
+      "near": 18
+    },
+    "inputRecords": 12214,
+    "nearDuplicateMethod": "word-trigram Jaccard after SimHash LSH candidate search",
+    "nearDuplicateThreshold": 0.92,
+    "retainedRecords": 12196,
+    "vagueEvalPolicy": "validation/test only"
+  },
+  "datasetVersion": "purpose-dataset-v1",
+  "frozenEval": {
+    "classifiableShippedFixtureRecords": 87,
+    "excludedGeneralFixtureRecords": 5,
+    "hardSliceDefinition": [
+      "boundary",
+      "mixed",
+      "pasted-context",
+      "vague-eval"
+    ],
+    "shippedFixtureRecords": 92,
+    "shippedFixturesPath": "Tests/NucleicCoreTests/Fixtures/purpose-prompts.json",
+    "shippedFixturesSha256": "6c8f0eedce35f9da7a407934009e87737b5e85c7371f6a118982d9fae2ea7aae",
+    "syntheticPath": "ml/purpose-classifier/data/frozen-test-v1.jsonl",
+    "syntheticSha256": "e5a6b501a43ce256092932ab05b0af01ff9137aee42d85b7d58bb4ce65f2943f"
+  },
+  "ratios": {
+    "test": 0.1,
+    "train": 0.8,
+    "validation": 0.1
+  },
+  "schemaVersion": 1,
+  "seed": 3248837105,
+  "sources": [
+    {
+      "path": "ml/purpose-classifier/data/purpose-prompts.jsonl",
+      "records": 9214,
+      "sha256": "3c7f8b496ef2014794a50433a78e3125c0dd4a03dc2a7773b52ee25587ac48e7"
+    },
+    {
+      "path": "ml/purpose-classifier/data/purpose-prompts-round2.jsonl",
+      "records": 3000,
+      "sha256": "0a35eac2549e95b31518cd0eee99b83b03761c26a0fb258f427fe04a632699ed"
+    }
+  ],
+  "splits": {
+    "test": {
+      "classifiableFixtureRecords": 87,
+      "distribution": {
+        "language": {
+          "de": 7,
+          "en": 1111,
+          "es": 9,
+          "fr": 5,
+          "ja": 3,
+          "pt": 4,
+          "zh": 3
+        },
+        "purpose": {
+          "backendImpl": 171,
+          "debugging": 149,
+          "frontendImpl": 157,
+          "planning": 132,
+          "quickFix": 140,
+          "refactor": 137,
+          "review": 127,
+          "writing": 129
+        },
+        "slice": {
+          "boundary": 179,
+          "core": 507,
+          "mixed": 80,
+          "pasted-context": 84,
+          "vague-eval": 292
+        }
+      },
+      "hardSyntheticRecords": 635,
+      "logicalRecords": 1229,
+      "sha256": "e5a6b501a43ce256092932ab05b0af01ff9137aee42d85b7d58bb4ce65f2943f",
+      "syntheticRecords": 1142
+    },
+    "train": {
+      "distribution": {
+        "language": {
+          "de": 100,
+          "en": 9326,
+          "es": 100,
+          "fr": 81,
+          "ja": 77,
+          "pt": 70,
+          "zh": 72
+        },
+        "purpose": {
+          "backendImpl": 1225,
+          "debugging": 1245,
+          "frontendImpl": 1235,
+          "planning": 1223,
+          "quickFix": 1230,
+          "refactor": 1225,
+          "review": 1230,
+          "writing": 1213
+        },
+        "slice": {
+          "boundary": 2065,
+          "core": 5820,
+          "mixed": 906,
+          "pasted-context": 1035
+        }
+      },
+      "records": 9826,
+      "sha256": "ca0ec9fe6fef6bc4013bf09bdfed5877327f69c98f8e5552c082cbb4e76d9b7d"
+    },
+    "validation": {
+      "distribution": {
+        "language": {
+          "de": 8,
+          "en": 1178,
+          "es": 12,
+          "fr": 7,
+          "ja": 8,
+          "pt": 8,
+          "zh": 7
+        },
+        "purpose": {
+          "backendImpl": 184,
+          "debugging": 157,
+          "frontendImpl": 168,
+          "planning": 141,
+          "quickFix": 151,
+          "refactor": 148,
+          "review": 140,
+          "writing": 139
+        },
+        "slice": {
+          "boundary": 196,
+          "core": 541,
+          "mixed": 83,
+          "pasted-context": 95,
+          "vague-eval": 313
+        }
+      },
+      "records": 1228,
+      "sha256": "7eca32f98105eec1c1a34c324851d52004d8dc720ca5853a1e91fa0a0dd620f4"
+    }
+  }
+}