|
Download analysis/README.md from SciKnowOrg/tabulus-runs: direct link, hf CLI and curl.
- Browser
- Download file 914 Bytes
-
https://huggingface.co/datasets/SciKnowOrg/tabulus-runs/resolve/main/analysis/README.md
- Command line
-
hf download hf://datasets/SciKnowOrg/tabulus-runs/analysis/README.md
-
curl -L -o README.md https://huggingface.co/datasets/SciKnowOrg/tabulus-runs/resolve/main/analysis/README.md
914 Bytes
Analysis summaries
This directory contains aggregate tables produced from the recorded Tabulus experiments.
all_runs.csv— consolidated run-level reconstruction/profiling summary. Cluster/session-specific identifiers and private filesystem paths have been removed; software versions, hardware model/resource quantities, timing, throughput, reconstruction yield, and outcome counts are retained.step5_runs.csv— aggregate Step 5 matching results by paper and adapter. The original internal output-path column has been removed.step5_ambiguous.csv— ambiguous Step 5 reference tokens.step5_skipped.csv— skipped Step 5 tables and reasons.step5_unmatched.csv— unmatched Step 5 reference tokens.
The historical stage field name in all_runs.csv is retained as part of the
recorded machine-generated table. Public package paths and documentation use
the term Step.