Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
a33e08d
1 Parent(s): 0fc4983

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -90,6 +90,24 @@ dataset_info:
90
  num_examples: 2199
91
  download_size: 70862028
92
  dataset_size: 73213942.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
93
  - config_name: vqarad
94
  features:
95
  - name: images
@@ -147,6 +165,10 @@ configs:
147
  data_files:
148
  - split: train
149
  path: tat_qa/train-*
 
 
 
 
150
  - config_name: vqarad
151
  data_files:
152
  - split: train
 
90
  num_examples: 2199
91
  download_size: 70862028
92
  dataset_size: 73213942.0
93
+ - config_name: tqa
94
+ features:
95
+ - name: images
96
+ sequence: image
97
+ - name: texts
98
+ list:
99
+ - name: user
100
+ dtype: string
101
+ - name: assistant
102
+ dtype: string
103
+ - name: source
104
+ dtype: string
105
+ splits:
106
+ - name: train
107
+ num_bytes: 380346870.806369
108
+ num_examples: 1493
109
+ download_size: 378238311
110
+ dataset_size: 380346870.806369
111
  - config_name: vqarad
112
  features:
113
  - name: images
 
165
  data_files:
166
  - split: train
167
  path: tat_qa/train-*
168
+ - config_name: tqa
169
+ data_files:
170
+ - split: train
171
+ path: tqa/train-*
172
  - config_name: vqarad
173
  data_files:
174
  - split: train