Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
aeda9b7
1 Parent(s): b8d7db5

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -252,6 +252,24 @@ dataset_info:
252
  num_examples: 9969
253
  download_size: 386023352
254
  dataset_size: 541250281.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
255
  - config_name: vqarad
256
  features:
257
  - name: images
@@ -345,6 +363,10 @@ configs:
345
  data_files:
346
  - split: train
347
  path: vistext/train-*
 
 
 
 
348
  - config_name: vqarad
349
  data_files:
350
  - split: train
 
252
  num_examples: 9969
253
  download_size: 386023352
254
  dataset_size: 541250281.0
255
+ - config_name: visualmrc
256
+ features:
257
+ - name: images
258
+ sequence: image
259
+ - name: texts
260
+ list:
261
+ - name: user
262
+ dtype: string
263
+ - name: assistant
264
+ dtype: string
265
+ - name: source
266
+ dtype: string
267
+ splits:
268
+ - name: train
269
+ num_bytes: 2941051627.2639995
270
+ num_examples: 3027
271
+ download_size: 2912911810
272
+ dataset_size: 2941051627.2639995
273
  - config_name: vqarad
274
  features:
275
  - name: images
 
363
  data_files:
364
  - split: train
365
  path: vistext/train-*
366
+ - config_name: visualmrc
367
+ data_files:
368
+ - split: train
369
+ path: visualmrc/train-*
370
  - config_name: vqarad
371
  data_files:
372
  - split: train