Card: links and online demo section
Browse files
README.md
CHANGED
|
@@ -12,11 +12,10 @@ tags:
|
|
| 12 |
- cpu
|
| 13 |
---
|
| 14 |
# docparse: a very small document parser
|
| 15 |
-
|
| 16 |
-
- **Try it:** [huggingface.co/spaces/Kedhareswer/docparse](https://huggingface.co/spaces/Kedhareswer/docparse) (drop a PDF, image, Word, Excel, CSV or PowerPoint file)
|
| 17 |
-
- **Weights and code:** [huggingface.co/Kedhareswer/docparse](https://huggingface.co/Kedhareswer/docparse): `artifacts/docparse/ocr.pt` (6.0 MB), the `docparse` package, benchmark reports and tests
|
| 18 |
-
- **Source:** [github.com/Kedhareswer/ML-AI](https://github.com/Kedhareswer/ML-AI) (`reasoning_harness/`)
|
| 19 |
-
|
| 20 |
|
| 21 |
`docparse` extracts **everything** from a document (text, headings, paragraphs,
|
| 22 |
lists, tables, sheets, formulas, headers/footers, metadata and figure positions)
|
|
@@ -257,6 +256,19 @@ Each block has a `type` (`heading`, `paragraph`, `list_item`, `table`,
|
|
| 257 |
an OCR `confidence`. Excel sheets also list formulas, merged ranges, charts and
|
| 258 |
images under `sheets`.
|
| 259 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 260 |
## Train or fine-tune your own
|
| 261 |
|
| 262 |
```bash
|
|
|
|
| 12 |
- cpu
|
| 13 |
---
|
| 14 |
# docparse: a very small document parser
|
| 15 |
+
|
| 16 |
+
- **Try it:** [huggingface.co/spaces/Kedhareswer/docparse](https://huggingface.co/spaces/Kedhareswer/docparse) (drop a PDF, image, Word, Excel, CSV or PowerPoint file)
|
| 17 |
+
- **Weights and code:** [huggingface.co/Kedhareswer/docparse](https://huggingface.co/Kedhareswer/docparse): `artifacts/docparse/ocr.pt` (6.0 MB), the `docparse` package, benchmark reports and tests
|
| 18 |
+
- **Source:** [github.com/Kedhareswer/ML-AI](https://github.com/Kedhareswer/ML-AI) (`reasoning_harness/`)
|
|
|
|
| 19 |
|
| 20 |
`docparse` extracts **everything** from a document (text, headings, paragraphs,
|
| 21 |
lists, tables, sheets, formulas, headers/footers, metadata and figure positions)
|
|
|
|
| 256 |
an OCR `confidence`. Excel sheets also list formulas, merged ranges, charts and
|
| 257 |
images under `sheets`.
|
| 258 |
|
| 259 |
+
## Online demo
|
| 260 |
+
|
| 261 |
+
The [Space](https://huggingface.co/spaces/Kedhareswer/docparse) runs the same
|
| 262 |
+
`docparse ui` server and weights. Files up to 40 MB, one at a time. LibreOffice
|
| 263 |
+
is installed there, so `.doc`, `.rtf`, `.odt`, `.xls`, `.ods`, `.ppt`, `.pptx`
|
| 264 |
+
and `.odp` work too. It runs on Hugging Face's free ZeroGPU hardware with all
|
| 265 |
+
inference on CPU (the app registers an unused `@spaces.GPU` function because
|
| 266 |
+
ZeroGPU will not start without one). Measured there: Word, Excel and CSV files
|
| 267 |
+
about 1 s, an 11-page text-layer PDF 3.5 s, a 14-page slide PDF 13 s, a page
|
| 268 |
+
image read by OCR 2.5 s. Uploaded files are written to a temporary folder for
|
| 269 |
+
the extraction and deleted right after. The Space build is
|
| 270 |
+
`scripts/make_hf_docparse.sh`.
|
| 271 |
+
|
| 272 |
## Train or fine-tune your own
|
| 273 |
|
| 274 |
```bash
|