Record full local runtime setup

This commit is contained in:
NINI
2026-05-10 01:39:07 +09:00
parent 4b316ebd0b
commit 80fda47163
2 changed files with 11 additions and 6 deletions
+1 -1
View File
@@ -4,7 +4,7 @@ This file is the shared work plan for agents. Read it before starting work, then
## Current Goal ## Current Goal
Completed work history is archived in `docs/WORKARCHIVE.md`. Sprint 10 pre-conversion PDF chunking is implemented. On this PC, base uv/Python project setup is complete in `.venv`; next runtime work is local MinerU/PyTorch/model/MathJax setup if requested. Completed work history is archived in `docs/WORKARCHIVE.md`. Sprint 10 pre-conversion PDF chunking is implemented. On this PC, full local runtime setup is complete in `.venv`; next work is optional manual Obsidian quality review or additional sample validation if requested.
## Active Constraints ## Active Constraints
+10 -5
View File
@@ -25,8 +25,9 @@ This file records current progress for agents. Read it before starting work, the
- MinerU execution mode: direct local `mineru` CLI only. - MinerU execution mode: direct local `mineru` CLI only.
- Strict-local allows MinerU 3.1.0's CLI-internal temporary local `mineru-api` when the CLI runs without `--api-url`. - Strict-local allows MinerU 3.1.0's CLI-internal temporary local `mineru-api` when the CLI runs without `--api-url`.
- Strict-local prohibits `--api-url`, remote APIs, router mode, HTTP client backends, and remote OpenAI-compatible backends. - Strict-local prohibits `--api-url`, remote APIs, router mode, HTTP client backends, and remote OpenAI-compatible backends.
- Current base `.venv` has project fast-test dependencies from `uv sync`. - Current `.venv` has project fast-test dependencies, CUDA-enabled PyTorch `2.6.0+cu126`, `torchvision 0.21.0+cu126`, and `mineru[core]==3.1.0`.
- Current `pdf2md doctor` status is FAIL because MinerU CLI is not installed on this PC. PyTorch, local MinerU model/cache paths, and local MathJax package setup are also not present yet. - Current `pdf2md doctor` status is PASS. MinerU, RTX 4080 SUPER CUDA PyTorch, local model config, MathJax, and strict-local checks pass.
- MinerU models were downloaded from Hugging Face by explicit setup command. Runtime model loading uses `MINERU_MODEL_SOURCE=local`.
## Recent Completed Work ## Recent Completed Work
@@ -39,6 +40,10 @@ This file records current progress for agents. Read it before starting work, the
- The FourNode sample conversion report status was `success`: 7 pages, 22 assets, 38 inline formulas, 16 display formulas, 0 math render errors, and 0 warnings. - The FourNode sample conversion report status was `success`: 7 pages, 22 assets, 38 inline formulas, 16 display formulas, 0 math render errors, and 0 warnings.
- Installed uv `0.11.12` at `C:\Users\baram\.local\bin`, installed uv-managed CPython `3.12.13`, created `.venv`, and ran `uv sync`. - Installed uv `0.11.12` at `C:\Users\baram\.local\bin`, installed uv-managed CPython `3.12.13`, created `.venv`, and ran `uv sync`.
- Verified base project environment with `uv run pytest`: 163 passed, 1 skipped. - Verified base project environment with `uv run pytest`: 163 passed, 1 skipped.
- Installed runtime dependencies on this PC: CUDA PyTorch `2.6.0+cu126`, `torchvision 0.21.0+cu126`, `mineru[core]==3.1.0`, local MathJax npm dependencies, and local MinerU models.
- Set user environment variable `MINERU_MODEL_SOURCE=local`.
- Verified full local runtime with `uv run pdf2md doctor`: PASS.
- Verified real local sample conversion: `samples/FourNodeQuadrilateralShellElementMITC4.pdf` to ignored `outputs/runtime-smoke/`, status `success`, 7 pages, 22 assets, 38 inline formulas, 16 display formulas, 0 math render errors, and 0 warnings.
## In Progress ## In Progress
@@ -46,10 +51,10 @@ This file records current progress for agents. Read it before starting work, the
## Blockers ## Blockers
- Real local MinerU conversion is blocked until MinerU 3.1.0, CUDA-enabled PyTorch, local models/cache, and optional MathJax package setup are installed on this PC. - No active blocker.
## Next Actions ## Next Actions
1. Install local MinerU 3.1.0 runtime stack on this PC if real conversions are needed. 1. Review generated sample Markdown outputs in Obsidian if visual quality needs manual assessment.
2. Install optional local MathJax dependencies with `npm install` if formula renderability checks should run. 2. Run optional real local chunked conversion on a long sample only if requested.
3. Preserve strict-local runtime behavior: use local model paths, direct CLI execution, and no user-specified API or remote backend. 3. Preserve strict-local runtime behavior: use local model paths, direct CLI execution, and no user-specified API or remote backend.