Datasets:

Modalities:
Image
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
Molbap HF staff commited on
Commit
b664b2f
1 Parent(s): 54b24ab

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -99,6 +99,20 @@ next(iter(dataset["train"])).keys()
99
  ```
100
  `image` will be a byte string containing the image contents. `answers` is a list of possible answers, aligned with the expected inputs to the [ANLS metric](https://arxiv.org/abs/1905.13648).
101
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
102
  The loader can then be iterated on normally and yields questions. Many questions rely on the same image, so there is some amount of data duplication.
103
 
104
  ### Data Splits
 
99
  ```
100
  `image` will be a byte string containing the image contents. `answers` is a list of possible answers, aligned with the expected inputs to the [ANLS metric](https://arxiv.org/abs/1905.13648).
101
 
102
+ Calling
103
+
104
+ ```python
105
+ from PIL import Image
106
+ from io import BytesIO
107
+ image = Image.open(BytesIO(docvqa_dataset["train"][0]["image"]['bytes']))
108
+ ```
109
+ will yield the image
110
+
111
+ <center>
112
+ <img src="https://huggingface.co/datasets/pixparse/docvqa-single-page-questions/resolve/main/doc_images_docvqa/docvqa_example.png" alt="An example of document available in docVQA" width="600" height="300">
113
+ <p><em>A document overlapping with tobacco on which questions are asked such as 'When is the contract effective date?' with the answer ['7 - 1 - 99']</em></p>
114
+ </center>
115
+
116
  The loader can then be iterated on normally and yields questions. Many questions rely on the same image, so there is some amount of data duplication.
117
 
118
  ### Data Splits