unstructured/test_unstructured_ingest/metrics/text-extraction/aggregate-scores-cct.tsv

strategy	average	sample_sd	population_sd	count
cct-accuracy	0.792	0.253	0.245	15
cct-%missing	0.025	0.034	0.033	15
build: text extraction evaluation metrics workflow added (#1757) Executive Summary This PR adds the evaluation metrics to our current workflow. It verifies the flow that when the code is pushed, the code will gets evaluate against our gold standard and output into `.tsv` file. Technical Details - Adds evaluation metrics to the test-ingest workflow - Make use of `structured-output` from `test-ingest` and compare to the gold-standard uploaded in s3, and download into local when make comparison. The current folder in-use is `s3://utic-dev-tech-fixtures/small-cct`. This dir is editable in the shell script. - With this PR, only one file from one connector is use to compare. Misc - Not many overlapped files between test-ingest and gold-standard. More files will be added. Outputs 2 `.tsv` files are saved under `test_unstructured_ingest/metrics/`. ![image](https://github.com/Unstructured-IO/unstructured/assets/2177850/222e437c-1a94-4d7c-9320-81696633b1ae) ![image](https://github.com/Unstructured-IO/unstructured/assets/2177850/5c840322-6739-4634-8868-eba04b4ebc96) --------- Co-authored-by: ryannikolaidis <1208590+ryannikolaidis@users.noreply.github.com> Co-authored-by: Klaijan <Klaijan@users.noreply.github.com> 2023-10-23 17:39:22 -04:00			`strategy average sample_sd population_sd count`
chore: get eval metrics from ingest in (#2097) Co-authored-by: ryannikolaidis <1208590+ryannikolaidis@users.noreply.github.com> Co-authored-by: Klaijan <Klaijan@users.noreply.github.com> 2023-11-17 13:22:36 -05:00			`cct-accuracy 0.792 0.253 0.245 15`
			`cct-%missing 0.025 0.034 0.033 15`