blood-test-explainer / docs /MESSAGE_TO_DIMITRIS.md
Codex
Default extraction to the fine-tuned MiniCPM-V Hub checkpoint.
c1a5fb4
|
Raw
History Blame Contribute Delete
1.25 kB
# Message to Dimitris (copy-paste)
Hi Dimitris β€” here’s what’s left before submission. Full checklist for you and your agents: **`docs/REMAINING_WORK.md`**.
**Priority order:** custom model β†’ copy pass β†’ KB + videos (agents) β†’ article β†’ demo video.
1. **Custom model** β€” Fine-tuned repo is live on Hub and wired as the code default; confirm Space after redeploy, run `modal run train/modal_eval.py::compare` for before/after numbers.
2. **Copy** β€” Tighten hero, trace steps, and README; fix badge claims (Well-Tuned only after model swap).
3. **Knowledge graph** β€” Expand beyond 107 markers via `markers.py` + `expand_lab_knowledge_graph.py` (agent task).
4. **Videos** β€” Replace reused YouTube URLs with marker/category-specific explainers (agent task).
5. **Article** β€” Problem β†’ architecture β†’ fine-tune proof β†’ limitations + Space link.
6. **Demo video** β€” Laytimely-style: AI voice + screen recording + background music; record after 1–2 are done.
**Current baseline:** App defaults to `build-small-hackathon/blood-test-minicpmv-4_6-medreason`; KG has 107 markers; only 2/13 real eval reports labeled; article and demo not started.
Ping me when the model is on the Space or if you want me on copy/KB review.