Datasets:

Languages:
English
Multilinguality:
monolingual
Language Creators:
machine-generated
Annotations Creators:
machine-generated
Source Datasets:
original
ArXiv:
Tags:
chained-qa
License:
system HF staff commited on
Commit
ab3777b
1 Parent(s): b1e4001

Update files from the datasets library (from 1.9.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.9.0

Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -419,7 +419,7 @@ The "types" are are:
419
  - `en-valid` and `en-valid-10k`
420
  - are the same as `en` and `en10k` except the train sets have been conveniently split into train and valid portions (90% and 10% split).
421
 
422
- To get a particular dataset, use `load_datasets('babi_qa',type=f'{type}',task_no=f'{task_no}')` where `type` is one of the types, and `task_no` is one of the task numbers. For example, `load_dataset('babi_qa', type='en', task_no='qa1')`.
423
  ### Languages
424
 
425
 
 
419
  - `en-valid` and `en-valid-10k`
420
  - are the same as `en` and `en10k` except the train sets have been conveniently split into train and valid portions (90% and 10% split).
421
 
422
+ To get a particular dataset, use `load_dataset('babi_qa',type=f'{type}',task_no=f'{task_no}')` where `type` is one of the types, and `task_no` is one of the task numbers. For example, `load_dataset('babi_qa', type='en', task_no='qa1')`.
423
  ### Languages
424
 
425