Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
HugoLaurencon commited on
Commit
6f833eb
1 Parent(s): 4c0dba4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -88,6 +88,20 @@ There is only one split, `train`, that contains 141,047,697 examples.
88
 
89
  `OBELISC` with images replaced by their URLs weighs 666.6 GB (unwanted!) in arrow format and 377 GB in this uploaded `parquet` format.
90
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
91
  ### Visualization of OBELISC documents
92
 
93
  https://huggingface.co/spaces/HuggingFaceM4/obelisc_visualization
 
88
 
89
  `OBELISC` with images replaced by their URLs weighs 666.6 GB (unwanted!) in arrow format and 377 GB in this uploaded `parquet` format.
90
 
91
+ ## Configs
92
+
93
+ The default config, downloaded when nothing is specified in the config argument, with
94
+ ```
95
+ from datasets import load_dataset
96
+
97
+ ds = load_dataset("HuggingFaceM4/OBELISC")
98
+ ```
99
+ corresponds to the original version of the dataset.
100
+
101
+ When building the dataset, we sent every image URL to the Spawning AI API and removed all the opted-out images.
102
+ However, we noticed afterward that some images might not be opted-out, but the whole web page containing them is.
103
+ This is why we created another config of the dataset to additionally filter out opted-out web pages, that can be loaded with `ds = load_dataset("HuggingFaceM4/OBELISC", config_name="opt_out_docs_removed")`.
104
+
105
  ### Visualization of OBELISC documents
106
 
107
  https://huggingface.co/spaces/HuggingFaceM4/obelisc_visualization