metadata
languages:
- en
task_categories:
- text-classification
AutoNLP Dataset for project: devign_raw_test
Dataset Descritpion
This dataset has been automatically processed by AutoNLP for project devign_raw_test.
Languages
The BCP-47 code for the dataset's language is en.
Dataset Structure
Data Instances
A sample from this dataset looks as follows:
[
{
"text": "void ff_avg_h264_qpel16_mc32_msa ( uint8_t * dst , const uint8_t * src , ptrdiff_t stride ) { avc_lu[...]",
"target": 0
},
{
"text": "static void sd_cardchange ( void * opaque , bool load ) { SDState * sd = opaque ; qemu_set_irq ( sd [...]",
"target": 0
}
]
Dataset Fields
The dataset has the following fields (also called "features"):
{
"text": "Value(dtype='string', id=None)",
"target": "ClassLabel(num_classes=2, names=['0', '1'], 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 | 21188 |
| valid | 5298 |