Commands cheat sheet

High-frequency entry points — full detail lives in Usage

Reference

High-frequency CLI entry points. Full flag semantics and recipes live in Usage.

Canonical reference

This page is a quick map. Flag semantics, flags, and recipes: Usage.

Environment

pip install -e ".[dev]"
python scripts/setup_env.py
python scripts/setup_env.py --status

DICIE (`src/docie`)

# one-shot text demo
python -m src.docie \
  --application salvage_claims \
  --text "LETTER OF GUARANTEE ..." \
  --response-only

# batch JSONL
python -m src.docie \
  --application medical_bills \
  --in data/eval/docie_eval_set.jsonl \
  --out-dir data/pipeline/docie_run

# optional REST surface
python -m src.docie.serve

Guides: DICIE Pipeline · Notebook

Memo chain (`src/pipeline`)

python -m src.pipeline.orchestrator \
  --in data/synthetic/documents/documents.jsonl \
  --out data/pipeline/analysis.jsonl \
  --vision

python -m src.pipeline.orchestrator --pdf claim.pdf --vision
python -m src.pipeline.batch_runner \
  --in data/synthetic/documents/documents.jsonl \
  --out-dir data/pipeline/batch_run \
  --vision

Guides: Architecture · Notebook

Sample corpus (`src/storage`)

python -m src.storage seed --seed 42 --also-export
python -m src.storage stats
python -m src.storage export-docie --application salvage_claims \
  --out data/sample_corpus/exports/salvage.jsonl

Guides: Sample Document Corpus · Walkthrough · SQL · Train/test

RVL-CDIP SQL (`src/rvl_cdip`)

python -m src.rvl_cdip build          # labels → .venv SQLite (~17 MB download)
python -m src.rvl_cdip summary
python -m src.rvl_cdip list --split train --label invoice --limit 5
python -m src.rvl_cdip query "SELECT split, COUNT(*) AS n FROM documents GROUP BY split"
# python -m src.rvl_cdip download-images --preflight
# python -m src.rvl_cdip download-images \
#   --i-understand-large-download --confirm-writes-under-venv   # ~38 GB → .venv only

Guide: RVL-CDIP SQL Index · OpenRouter eval notebook

Classification & extraction (training)

python -m src.classification.prepare_dataset \
  --in data/synthetic/documents/documents_from_skeletons_n240_seed42.jsonl
python -m src.classification.train_classifier \
  --prepared data/synthetic/documents/classification_prepared --smoke
python -m src.classification.eval \
  --model-dir models/classifier_smoke \
  --prepared data/synthetic/documents/classification_prepared

Classical baseline notebook: Random Forest

Discord bot

pip install -e ".[discord]"
python -m src.discord_bot

Documentation site

./scripts/preview_docs_site.sh          # local Quarto preview
./scripts/publish_docs_site.sh          # Posit Connect Cloud

--- title: "Commands cheat sheet" subtitle: "High-frequency entry points — full detail lives in Usage" --- ::: {.hero-kicker} Reference ::: ::: {.hero-lead} High-frequency CLI entry points. Full flag semantics and recipes live in [Usage](../usage.md). ::: ::: {.callout-note} ## Canonical reference This page is a **quick map**. Flag semantics, flags, and recipes: [Usage](../usage.md). ::: ## Environment ```bash pip install -e ".[dev]" python scripts/setup_env.py python scripts/setup_env.py --status ``` ## DICIE (`src/docie`) ```bash # one-shot text demo python -m src.docie \ --application salvage_claims \ --text "LETTER OF GUARANTEE ..." \ --response-only # batch JSONL python -m src.docie \ --application medical_bills \ --in data/eval/docie_eval_set.jsonl \ --out-dir data/pipeline/docie_run # optional REST surface python -m src.docie.serve ``` Guides: [DICIE Pipeline](../docie_pipeline.qmd) · [Notebook](../notebooks/docie_pipeline_walkthrough.ipynb) ## Memo chain (`src/pipeline`) ```bash python -m src.pipeline.orchestrator \ --in data/synthetic/documents/documents.jsonl \ --out data/pipeline/analysis.jsonl \ --vision python -m src.pipeline.orchestrator --pdf claim.pdf --vision python -m src.pipeline.batch_runner \ --in data/synthetic/documents/documents.jsonl \ --out-dir data/pipeline/batch_run \ --vision ``` Guides: [Architecture](../architecture.qmd) · [Notebook](../notebooks/pipeline_walkthrough.ipynb) ## Sample corpus (`src/storage`) ```bash python -m src.storage seed --seed 42 --also-export python -m src.storage stats python -m src.storage export-docie --application salvage_claims \ --out data/sample_corpus/exports/salvage.jsonl ``` Guides: [Sample Document Corpus](../sample_document_corpus.md) · [Walkthrough](../notebooks/sample_document_corpus_walkthrough.ipynb) · [SQL](../notebooks/sample_corpus_sql_integrations.ipynb) · [Train/test](../notebooks/sample_corpus_train_test_pipeline.ipynb) ## RVL-CDIP SQL (`src/rvl_cdip`) ```bash python -m src.rvl_cdip build # labels → .venv SQLite (~17 MB download) python -m src.rvl_cdip summary python -m src.rvl_cdip list --split train --label invoice --limit 5 python -m src.rvl_cdip query "SELECT split, COUNT(*) AS n FROM documents GROUP BY split" # python -m src.rvl_cdip download-images --preflight # python -m src.rvl_cdip download-images \ # --i-understand-large-download --confirm-writes-under-venv # ~38 GB → .venv only ``` Guide: [RVL-CDIP SQL Index](../rvl_cdip_sql.md) · [OpenRouter eval notebook](../notebooks/rvl_cdip_openrouter_eval.ipynb) ## Classification & extraction (training) ```bash python -m src.classification.prepare_dataset \ --in data/synthetic/documents/documents_from_skeletons_n240_seed42.jsonl python -m src.classification.train_classifier \ --prepared data/synthetic/documents/classification_prepared --smoke python -m src.classification.eval \ --model-dir models/classifier_smoke \ --prepared data/synthetic/documents/classification_prepared ``` Classical baseline notebook: [Random Forest](../notebooks/random_forest_text_handwriting_classification.ipynb) ## Discord bot ```bash pip install -e ".[discord]" python -m src.discord_bot ``` ## Documentation site ```bash ./scripts/preview_docs_site.sh # local Quarto preview ./scripts/publish_docs_site.sh # Posit Connect Cloud ``` ::: {.see-also} ### See also [Quick Start](../quick-start.qmd) · [Usage](../usage.md) · [Pipeline hub](../pipelines/index.qmd) · [How-to: Quarto site](../how-to/launch-quarto-site.qmd) :::

Environment

DICIE (src/docie)

Memo chain (src/pipeline)

Sample corpus (src/storage)

RVL-CDIP SQL (src/rvl_cdip)

Classification & extraction (training)

Discord bot

Documentation site

See also

DICIE (`src/docie`)

Memo chain (`src/pipeline`)

Sample corpus (`src/storage`)

RVL-CDIP SQL (`src/rvl_cdip`)