Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
fdc9653
1 Parent(s): 59e5795

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -72,6 +72,24 @@ dataset_info:
72
  num_examples: 18266
73
  download_size: 803229473
74
  dataset_size: 784762327.9534782
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75
  - config_name: datikz
76
  features:
77
  - name: images
@@ -557,6 +575,10 @@ configs:
557
  data_files:
558
  - split: train
559
  path: chartqa/train-*
 
 
 
 
560
  - config_name: datikz
561
  data_files:
562
  - split: train
 
72
  num_examples: 18266
73
  download_size: 803229473
74
  dataset_size: 784762327.9534782
75
+ - config_name: cocoqa
76
+ features:
77
+ - name: images
78
+ sequence: image
79
+ - name: texts
80
+ list:
81
+ - name: user
82
+ dtype: string
83
+ - name: assistant
84
+ dtype: string
85
+ - name: source
86
+ dtype: string
87
+ splits:
88
+ - name: train
89
+ num_bytes: 2213960474.0
90
+ num_examples: 46287
91
+ download_size: 2393991009
92
+ dataset_size: 2213960474.0
93
  - config_name: datikz
94
  features:
95
  - name: images
 
575
  data_files:
576
  - split: train
577
  path: chartqa/train-*
578
+ - config_name: cocoqa
579
+ data_files:
580
+ - split: train
581
+ path: cocoqa/train-*
582
  - config_name: datikz
583
  data_files:
584
  - split: train