CoNaLa / README.md
autonlp-data-processor
Processed data from autonlp data processor ([2021-10-18 03:51 ]
9a02dd2
|
raw
history blame
1.27 kB
---
task_categories:
- conditional-text-generation
---
# AutoNLP Dataset for project: Code_generation
## Table of content
- [Dataset Description](#dataset-description)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
## Dataset Descritpion
This dataset has been automatically processed by AutoNLP for project Code_generation.
### Languages
The BCP-47 code for the dataset's language is unk.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"text": "convert a list to a dictionary in python",
"target": "b = dict(zip(a[0::2], a[1::2]))"
},
{
"text": "python - sort a list of nested lists",
"target": "l.sort(key=sum_nested)"
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"text": "Value(dtype='string', id=None)",
"target": "Value(dtype='string', id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 11125 |
| valid | 1237 |