File size: 695 Bytes
0691b0e 7483ae6 0691b0e 7483ae6 0691b0e 7483ae6 0691b0e 6dfb4c2 0691b0e 6dfb4c2 795a601 6dfb4c2 795a601 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
---
dataset_info:
features:
- name: incorrect
dtype: string
- name: correct
dtype: string
splits:
- name: train
num_bytes: 193960468
num_examples: 540000
- name: test
num_bytes: 21546211
num_examples: 60000
download_size: 129134550
dataset_size: 215506679
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
language:
- ur
---
This dataset contains paired Urdu text for error correction tasks.
## Data Format
Each dialogue entry in the dataset is structured as follows:
- `input_text`: The text containing simulated errors.
- `correct_text`: The corresponding correct text.
|