You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-**Discovery accessible from CLI and SDK** — Discovery can now be run programmatically via the IDP SDK (`client.discovery.run()`) and CLI (`idp-cli discover`), enabling users with many document classes to automate schema generation without the Web UI. Supports both modes: without ground truth (exploratory) and with ground truth (optimized). ([#228](https://github.com/aws-solutions-library-samples/accelerated-intelligent-document-processing-on-aws/issues/228))
206
206
207
-
-**Custom Model Fine-tuning** — Fine-tune Amazon Nova models using processed IDP documents directly from the Web UI. Create a fine-tuning job from a Test Set (which provides documents with ground truth data), generate training data in JSONL format, submit fine-tuning jobs to Bedrock, and deploy the resulting custom models for use in extraction workflows. See [Custom Model Fine-tuning](./docs/custom-model-finetuning.md) for details.
207
+
-**Custom Model Fine-tuning** — Improve extraction and classification accuracy for your specific document types by fine-tuning Amazon Nova models on your own labeled data — no ML expertise required. Select a Test Set with ground truth, choose a base model, and the system handles training data generation, Bedrock fine-tuning, and on-demand model deployment automatically. Custom models are billed pay-per-token with no idle costs. Available to Admin and Author roles. See [Custom Model Fine-tuning](./docs/custom-model-finetuning.md) for details.
208
208
-**Web UI**: New "Custom Models" page with job creation form (test set selector, base model selector, train/validation split), jobs table with status tracking, and detailed job view with deployment status and configuration version creation
-**GraphQL API**: New `createFinetuningJob`, `getFinetuningJob`, `listFinetuningJobs`, `deleteFinetuningJob` mutations/queries with `FinetuningJob` type and real-time status fields
Copy file name to clipboardExpand all lines: docs/custom-model-finetuning.md
+7Lines changed: 7 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -28,13 +28,20 @@ flowchart LR
28
28
5.**Deploy Endpoint**: The trained model is deployed as an on-demand Custom Model Deployment
29
29
6.**Use in Configuration**: Select the custom model when creating or editing configurations
30
30
31
+
## Roles & Permissions
32
+
33
+
Custom Model Fine-tuning is available to **Admin** and **Author** roles. The GraphQL mutations (`createFinetuningJob`, `deleteFinetuningJob`) are enforced server-side via AppSync `@aws_auth` directives, so Reviewer and Viewer roles cannot create or delete jobs even via direct API calls. The query APIs (`listFinetuningJobs`, `getFinetuningJob`, `listAvailableModels`) are accessible to all authenticated users at the schema level, but the UI only exposes the Custom Models page to Admin and Author.
34
+
35
+
> **Note on config-version scoping**: Fine-tuning jobs are currently global — they are not filtered by `allowedConfigVersions`. A scoped Author can see all fine-tuning jobs and create jobs from any test set. However, when applying a custom model to a configuration version (via "Create Config Version"), the config-version scope is enforced — the Author can only target versions within their scope. See [RBAC](./rbac.md) for the full permission matrix.
36
+
31
37
## Prerequisites
32
38
33
39
- An existing Test Set with:
34
40
- At least 10 documents
35
41
- At least 2 document classes
36
42
- Baseline classifications for all documents
37
43
- Appropriate IAM permissions for Bedrock model customization
44
+
-**Admin** or **Author** role in the IDP Accelerator
|`listFinetuningJobs`, `getFinetuningJob`, `validateTestSetForFinetuning`, `listAvailableModels`| All authenticated (UI limited to Admin, Author) |
199
207
|`queryKnowledgeBase`, `chatWithDocument`| All authenticated |
200
208
|`listUsers`| All authenticated (non-admin sees only self in resolver) |
201
209
|`getMyProfile`| All authenticated |
@@ -295,4 +303,5 @@ To add a new role:
295
303
-**Knowledge Base queries** do not currently enforce config-version scope. KB results may include documents from out-of-scope config versions.
296
304
-**Agent Companion Chat** analytics queries (Athena) do not filter by config-version scope.
297
305
-**GetDocument API** (direct document access by URL) does not enforce config-version scope at the resolver level. UI navigation hides out-of-scope documents, but direct API access is not blocked.
306
+
-**Custom Model Fine-tuning** jobs are global — not scoped by `allowedConfigVersions`. A scoped Author can see all fine-tuning jobs and create jobs from any test set. However, when applying a custom model to a configuration version (via the "Create Config Version" modal), the config-version scope IS enforced — the Author can only target versions within their scope.
298
307
- These limitations are tracked for Phase 3 implementation.
0 commit comments