Introduce v3 schema

Closes #38 v3 includes some backwards incompatible changes to the knowledge schema format. Here is a diff against v2. The changes are: - Q&A pairs now have an associated context blob from the knowledge document. - There is new `document_outline` field. - drop `task_description` ```diff --- src/instructlab/schema/v2/knowledge.json 2024-07-17 12:56:37 +++ src/instructlab/schema/v3/knowledge.json 2024-07-17 15:38:30 @@ -6,9 +6,9 @@ "required": [ "created_by", "domain", - "task_description", "seed_examples", - "document" + "document", + "document_outline" ], "unevaluatedProperties": false, "properties": { @@ -27,15 +27,6 @@ "Pop culture" ] }, - "task_description": { - "description": "A description of the task which is used in prompts to the teacher model during synthetic data generation. The description should be detailed and prescriptive to improve the teacher model's responses.", - "type": "string", - "minLength": 1, - "examples": [ - "To teach a language model about softball history", - "To teach a language model about tabby cats" - ] - }, "seed_examples": { "description": "An array of seed examples for synthetic data generation.", "type": "array", @@ -44,20 +35,39 @@ "items": { "type": "object", "required": [ - "question", - "answer" + "context", + "questions_and_answers" ], "unevaluatedProperties": false, "properties": { - "question": { - "description": "A question used for synthetic data generation.", + "context": { + "description": "Context from the document associated with this set of sample q&a pairs.", "type": "string", "minLength": 1 }, - "answer": { - "description": "The desired response for the question.", - "type": "string", - "minLength": 1 + "questions_and_answers": { + "type": "array", + "minItems": 3, + "uniqueItems": true, + "items": { + "type": "object", + "required": [ + "question", + "answer" + ], + "properties": { + "question": { + "description": "A question used for synthetic data generation.", + "type": "string", + "minLength": 1 + }, + "answer": { + "description": "The desired response for the question.", + "type": "string", + "minLength": 1 + } + } + } } } } @@ -104,6 +114,11 @@ } } } + }, + "document_outline": { + "description": "An outline of the document.", + "type": "string", + "minLength": 1 } } } ``` Signed-off-by: Russell Bryant <[email protected]>
instructlab · Jul 17, 2024 · 2be6938 · 2be6938
1 parent 859cb72
commit 2be6938
Show file tree

Hide file tree

Showing 3 changed files with 139 additions and 0 deletions.
diff --git a/src/instructlab/schema/v3/__init__.py b/src/instructlab/schema/v3/__init__.py
diff --git a/src/instructlab/schema/v3/knowledge.json b/src/instructlab/schema/v3/knowledge.json
@@ -0,0 +1,124 @@
+{
+    "title": "Knowledge",
+    "description": "A knowledge skill.",
+    "type": "object",
+    "$ref": "./version.json",
+    "required": [
+        "created_by",
+        "domain",
+        "seed_examples",
+        "document",
+        "document_outline"
+    ],
+    "unevaluatedProperties": false,
+    "properties": {
+        "created_by": {
+            "description": "The GitHub username of the contributor.",
+            "type": "string",
+            "minLength": 1
+        },
+        "domain": {
+            "description": "The knowledge domain which is used in prompts to the teacher model during synthetic data generation. The domain should be brief such as the title to a textbook chapter or section.",
+            "type": "string",
+            "minLength": 1,
+            "examples": [
+                "Chemistry",
+                "History",
+                "Pop culture"
+            ]
+        },
+        "seed_examples": {
+            "description": "An array of seed examples for synthetic data generation.",
+            "type": "array",
+            "minItems": 5,
+            "uniqueItems": true,
+            "items": {
+                "type": "object",
+                "required": [
+                    "context",
+                    "questions_and_answers"
+                ],
+                "unevaluatedProperties": false,
+                "properties": {
+                    "context": {
+                        "description": "Context from the document associated with this set of sample q&a pairs.",
+                        "type": "string",
+                        "minLength": 1
+                    },
+                    "questions_and_answers": {
+                        "type": "array",
+                        "minItems": 3,
+                        "uniqueItems": true,
+                        "items": {
+                            "type": "object",
+                            "required": [
+                                "question",
+                                "answer"
+                            ],
+                            "properties": {
+                                "question": {
+                                    "description": "A question used for synthetic data generation.",
+                                    "type": "string",
+                                    "minLength": 1
+                                },
+                                "answer": {
+                                    "description": "The desired response for the question.",
+                                    "type": "string",
+                                    "minLength": 1
+                                }
+                            }
+                        }
+                    }
+                }
+            }
+        },
+        "document": {
+            "description": "The knowledge documents.",
+            "type": "object",
+            "required": [
+                "repo",
+                "commit",
+                "patterns"
+            ],
+            "unevaluatedProperties": false,
+            "properties": {
+                "repo": {
+                    "description": "The URL to a Git repository holding the knowledge documents.",
+                    "type": "string",
+                    "minLength": 1,
+                    "examples": [
+                        "https://github.com/instructlab/instructlab.git"
+                    ]
+                },
+                "commit": {
+                    "description": "The commit in the Git repository containing the knowledge documents.",
+                    "type": "string",
+                    "minLength": 1,
+                    "examples": [
+                        "951999afdc59c46d325493568193b40bd5439c9e"
+                    ]
+                },
+                "patterns": {
+                    "description": "An array of glob patterns of the knowledge documents in the Git repository.",
+                    "type": "array",
+                    "minItems": 1,
+                    "uniqueItems": true,
+                    "items": {
+                        "type": "string",
+                        "minLength": 1,
+                        "examples": [
+                            "*.md",
+                            "folder/*.md",
+                            "folder/knowledge_doc.md"
+                        ]
+                    }
+                }
+            }
+        },
+        "document_outline": {
+            "description": "An outline of the document.",
+            "type": "string",
+            "minLength": 1
+        }
+    }
+}
diff --git a/src/instructlab/schema/v3/version.json b/src/instructlab/schema/v3/version.json
@@ -0,0 +1,15 @@
+{
+    "title": "Taxonomy Document Schema Version",
+    "type": "object",
+    "required": [
+        "version"
+    ],
+    "properties": {
+        "version": {
+            "description": "The schema version of the taxonomy document.",
+            "type": "integer",
+            "$comment": "This value must match the number in the containing folder.",
+            "const": 3
+        }
+    }
+}