Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
e64f1cf
1 Parent(s): b5b8cf1

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -720,6 +720,24 @@ dataset_info:
720
  num_examples: 313
721
  download_size: 16226241
722
  dataset_size: 16561537.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
723
  - config_name: vsr
724
  features:
725
  - name: images
@@ -899,6 +917,10 @@ configs:
899
  data_files:
900
  - split: train
901
  path: vqarad/train-*
 
 
 
 
902
  - config_name: vsr
903
  data_files:
904
  - split: train
 
720
  num_examples: 313
721
  download_size: 16226241
722
  dataset_size: 16561537.0
723
+ - config_name: vqav2
724
+ features:
725
+ - name: images
726
+ sequence: image
727
+ - name: texts
728
+ list:
729
+ - name: user
730
+ dtype: string
731
+ - name: assistant
732
+ dtype: string
733
+ - name: source
734
+ dtype: string
735
+ splits:
736
+ - name: train
737
+ num_bytes: 10630091683.0
738
+ num_examples: 82772
739
+ download_size: 13479302437
740
+ dataset_size: 10630091683.0
741
  - config_name: vsr
742
  features:
743
  - name: images
 
917
  data_files:
918
  - split: train
919
  path: vqarad/train-*
920
+ - config_name: vqav2
921
+ data_files:
922
+ - split: train
923
+ path: vqav2/train-*
924
  - config_name: vsr
925
  data_files:
926
  - split: train