unstructured/all-docs-cct.tsv at 466255eec36b2e57a02c4a1a458c110cd33cfda3

mirror of https://github.com/Unstructured-IO/unstructured.git synced 2025-06-27 02:30:08 +00:00

build: element type frequency evaluation metrics workflow in ci (#1862 )

**Executive Summary**
Measured element type frequency accuracy from the current version of
code with the expected output. The performance is reported as tsv file
under `metrics`.

**Technical Details**
- The evaluation measures element type frequencies from
`structured-output-eval` against `expected-structured-output`
- `evaluation.py` has been edited to support function calling using
`click.group()` and `command()`
- `evaluation-ingest-cp.sh` is now added to all the `test-ingest-xx.sh`
scripts

**Outputs**
2 tsv files is saved

![image](https://github.com/Unstructured-IO/unstructured/assets/2177850/b4458094-a9fc-48f9-a0bd-2ccd6985440a)

![image](https://github.com/Unstructured-IO/unstructured/assets/2177850/6d785736-bcaf-4275-bf2d-ab511cdfb3f4)
9-0e05-41d4-b69f-841a2aa131ec)
and aggregated score is displayed.

![image](https://github.com/Unstructured-IO/unstructured/assets/2177850/9d42bd0c-a0dd-41c2-a2e5-b675a40f35cc)

---------

Co-authored-by: ryannikolaidis <1208590+ryannikolaidis@users.noreply.github.com>
Co-authored-by: Klaijan <Klaijan@users.noreply.github.com>
Co-authored-by: Yao You <theyaoyou@gmail.com>

2023-10-27 04:36:36 +00:00

161 B

Raw Blame History

1	filename	connector	cct-accuracy	cct-%missing
2	IRS-form-1987.pdf	azure	0.783	0.13
3	example-10k.html	local	0.686	0.04
4	science-exploration-1p.pptx	box	0.861	0.09

161 B Raw Blame History

161 B

Raw Blame History