mirror of
https://github.com/MODSetter/SurfSense.git
synced 2026-05-17 18:35:19 +02:00
9bcd50164d
Adds the full parser_compare experiment for the multimodal_doc suite:
six arms compared on 30 PDFs / 171 questions from MMLongBench-Doc with
anthropic/claude-sonnet-4.5 across the board.
Source code:
- core/parsers/{azure_di,llamacloud,pdf_pages}.py: direct parser SDK
callers (Azure Document Intelligence prebuilt-read/layout, LlamaParse
parse_page_with_llm/parse_page_with_agent) used by the LC arms,
bypassing the SurfSense backend so each (basic/premium) extraction
is a clean A/B independent of backend ETL routing.
- suites/multimodal_doc/parser_compare/{ingest,runner,prompt}.py:
six-arm benchmark (native_pdf, azure_basic_lc, azure_premium_lc,
llamacloud_basic_lc, llamacloud_premium_lc, surfsense_agentic) with
byte-identical prompts per question, deterministic grader, Wilson
CIs, and the per-page preprocessing tariff cost overlay.
Reproducibility:
- pyproject.toml + uv.lock pin pypdf, azure-ai-documentintelligence,
llama-cloud-services as new deps.
- .env.example documents the AZURE_DI_* and LLAMA_CLOUD_API_KEY env
vars now required for parser_compare.
- 12 analysis scripts under scripts/: retry pass with exponential
backoff, post-retry accuracy merge, McNemar / latency / per-PDF
stats, context-overflow hypothesis test, etc. Each produces one
number cited by the blog report.
Citation surface:
- reports/blog/multimodal_doc_parser_compare_n171_report.md: 1219-line
technical writeup (16 sections) covering headline accuracy, per-format
accuracy, McNemar pairwise significance, latency / token / per-PDF
distributions, error analysis, retry experiment, post-retry final
accuracy, cost amortization model with closed-form derivation, threats
to validity, and reproducibility appendix.
- data/multimodal_doc/runs/2026-05-14T00-53-19Z/parser_compare/{raw,
raw_retries,raw_post_retry}.jsonl + run_artifact.json + retry summary
whitelisted via data/.gitignore as the verifiable numbers source.
Gitignore:
- ignore logs_*.txt + retry_run.log; structured artifacts cover the
citation surface, debug logs are noise.
- data/.gitignore default-ignores everything, whitelists the n=171 run
artifacts only (parser manifest left ignored to avoid leaking local
Windows usernames in absolute paths; manifest is fully regenerable
via 'ingest multimodal_doc parser_compare').
- reports/.gitignore now whitelists hand-curated reports/blog/.
Also retires the abandoned CRAG Task 3 implementation (download script,
streaming Task 3 ingest, CragTask3Benchmark + tests) and trims the
runner / ingest module APIs to match.
Co-authored-by: Cursor <cursoragent@cursor.com>
63 lines
1.5 KiB
TOML
63 lines
1.5 KiB
TOML
[project]
|
|
name = "surfsense-evals"
|
|
version = "0.1.0"
|
|
description = "Domain-agnostic evaluation harness for SurfSense (medical RAG suite ships first; legal/finance/code suites slot in under suites/)."
|
|
readme = "README.md"
|
|
requires-python = ">=3.12"
|
|
license = { text = "Apache-2.0" }
|
|
authors = [{ name = "SurfSense" }]
|
|
|
|
dependencies = [
|
|
"httpx>=0.27.0",
|
|
"httpx-sse>=0.4.0",
|
|
"datasets>=2.21.0",
|
|
"huggingface_hub>=0.24.0",
|
|
"reportlab>=4.0.0",
|
|
"Pillow>=10.0.0",
|
|
"pyarrow>=15.0.0",
|
|
"pydantic>=2.6.0",
|
|
"tqdm>=4.66.0",
|
|
"numpy>=1.26.0",
|
|
"scikit-learn>=1.4.0",
|
|
"scipy>=1.12.0",
|
|
"python-dotenv>=1.0.0",
|
|
"rich>=13.7.0",
|
|
"trafilatura>=1.12.0",
|
|
"pypdf>=5.1.0",
|
|
"azure-ai-documentintelligence>=1.0.2",
|
|
"llama-cloud-services>=0.6.25",
|
|
]
|
|
|
|
[project.optional-dependencies]
|
|
dev = [
|
|
"pytest>=8.0.0",
|
|
"pytest-asyncio>=0.23.0",
|
|
"respx>=0.21.0",
|
|
"ruff>=0.5.0",
|
|
]
|
|
|
|
[project.scripts]
|
|
surfsense-evals = "surfsense_evals.core.cli:main"
|
|
|
|
[build-system]
|
|
requires = ["setuptools>=61.0", "wheel"]
|
|
build-backend = "setuptools.build_meta"
|
|
|
|
[tool.setuptools.packages.find]
|
|
where = ["src"]
|
|
include = ["surfsense_evals*"]
|
|
|
|
[tool.pytest.ini_options]
|
|
asyncio_mode = "auto"
|
|
testpaths = ["tests"]
|
|
markers = [
|
|
"integration: opt-in tests that hit a live SurfSense instance (run with `-m integration`)",
|
|
]
|
|
|
|
[tool.ruff]
|
|
line-length = 100
|
|
target-version = "py312"
|
|
|
|
[tool.ruff.lint]
|
|
select = ["E", "F", "I", "B", "UP", "SIM", "ASYNC"]
|
|
ignore = ["E501"]
|