riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-02 19:21:31 +06:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

Adding doctest for document-question-answering (#20239)

Browse Source 
* Adding doctest for doc qa.

* Adding doctest for doc qa.

* Fixup.
This commit is contained in:
Nicolas Patry 2022-11-16 09:52:35 +01:00 committed by GitHub
parent 443aaaa1a7
commit 1f029b6ae7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 16 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
src/transformers/pipelines/document_question_answering.py
View File

@ -106,6 +106,22 @@ class DocumentQuestionAnsweringPipeline(ChunkPipeline):
similar to the (extractive) question answering pipeline; however, the pipeline takes an image (and optional OCR'd
words/boxes) as input instead of text context.
Example:
```python
>>> from transformers import pipeline
>>> document_qa = pipeline(model="impira/layoutlm-document-qa")
>>> result = document_qa(
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
... question="What is the invoice number?",
... )
>>> result[0]["answer"]
'1110212019'
```
[Using pipelines in a webserver or with a dataset](../pipeline_tutorial)
This document question answering pipeline can currently be loaded from [`pipeline`] using the following task
identifier: `"document-question-answering"`.