Skip to content

Add trainable Model primitive examples (IMDB NB + HF)#1

Merged
endymion merged 14 commits intomainfrom
feature/classification-examples
Feb 14, 2026
Merged

Add trainable Model primitive examples (IMDB NB + HF)#1
endymion merged 14 commits intomainfrom
feature/classification-examples

Conversation

@endymion
Copy link
Copy Markdown
Contributor

Adds two registry-backed, trainable Model primitive examples with deterministic mocks/specs (Naive Bayes + HF sequence classifier), updates existing LLM classification and HITL examples to use the current mocking story, and fixes the CI test runner Python selection/counter logic.

Related: Tactus-84l.3

- Add support inbox triage example with retry-focused BDD spec\n- Add composite fuzzy-then-LLM example with deterministic mock-mode specs\n- Include snippet markers for docs ingestion
- Add chapter 03-human-in-the-loop/ for HITL patterns
- 01-basic-hitl.tac: traditional HITL with direct primitive calls
- 02-agent-driven-approval.tac: agent-driven HITL with tool wrappers
- Demonstrates progression from hardcoded to intelligent HITL workflows
- Add support inbox triage example with retry-focused BDD spec\n- Add composite fuzzy-then-LLM example with deterministic mock-mode specs\n- Include snippet markers for docs ingestion
- Add chapter 03-human-in-the-loop/ for HITL patterns
- 01-basic-hitl.tac: traditional HITL with direct primitive calls
- 02-agent-driven-approval.tac: agent-driven HITL with tool wrappers
- Demonstrates progression from hardcoded to intelligent HITL workflows
@endymion endymion merged commit 19b840e into main Feb 14, 2026
2 checks passed
@endymion endymion deleted the feature/classification-examples branch February 14, 2026 09:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant