Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
d50c875
1 Parent(s): d6af7d2

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -126,6 +126,24 @@ dataset_info:
126
  num_examples: 47974
127
  download_size: 613100257
128
  dataset_size: 481233278.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
129
  - config_name: docvqa
130
  features:
131
  - name: images
@@ -839,6 +857,10 @@ configs:
839
  data_files:
840
  - split: train
841
  path: datikz/train-*
 
 
 
 
842
  - config_name: docvqa
843
  data_files:
844
  - split: train
 
126
  num_examples: 47974
127
  download_size: 613100257
128
  dataset_size: 481233278.0
129
+ - config_name: diagram_image_to_text
130
+ features:
131
+ - name: images
132
+ sequence: image
133
+ - name: texts
134
+ list:
135
+ - name: user
136
+ dtype: string
137
+ - name: assistant
138
+ dtype: string
139
+ - name: source
140
+ dtype: string
141
+ splits:
142
+ - name: train
143
+ num_bytes: 18877197.0
144
+ num_examples: 300
145
+ download_size: 18706661
146
+ dataset_size: 18877197.0
147
  - config_name: docvqa
148
  features:
149
  - name: images
 
857
  data_files:
858
  - split: train
859
  path: datikz/train-*
860
+ - config_name: diagram_image_to_text
861
+ data_files:
862
+ - split: train
863
+ path: diagram_image_to_text/train-*
864
  - config_name: docvqa
865
  data_files:
866
  - split: train