metadata
datasets:
- code-search-net/code_search_net
language:
- en
base_model:
- openai-community/gpt2
pipeline_tag: text-generation
library_name: transformers
tags:
- code
Detailed Model Description
A GPT-2-based tokenizer further trained on 400 k+ Python functions. It keeps the original BPE backbone, adds robust encoding for indentation, common keywords, operators and camel-case variables, and is ready for any code-generation or code-understanding pipeline.
Usage Examples:
example = """
class LinearLayer():
def __init__(self, input_size, output_size):
self.weight = torch.randn(input_size, output_size)
self.bias = torch.zeros(output_size)
def __call__(self, x):
return x @ self.weights + self.bias
"""
Performance：
['class', 'ĠLinear', 'Layer', '():', 'ĊĠĠĠ', 'Ġdef', 'Ġ__', 'init', '__(', 'self', ',', 'Ġinput', '_', 'size', ',', 'Ġoutput', '_', 'size', '):', 'ĊĠĠĠĠĠĠĠ', 'Ġself', '.', 'weight', 'Ġ=', 'Ġtorch', '.', 'randn', '(', 'input', '_', 'size', ',', 'Ġoutput', '_', 'size', ')', 'ĊĠĠĠĠĠĠ', 'Ġself', '.', 'bias', 'Ġ=', 'Ġtorch', '.', 'zeros', '(', 'output', '_', 'size', ')', 'ĊĊĠĠ', 'Ġdef', 'Ġ__', 'call', '__(', 'self', ',', 'Ġx', '):', 'ĊĠĠĠĠĠĠ', 'Ġreturn', 'Ġx', 'Ġ@', 'Ġself', '.', 'weights', 'Ġ+', 'Ġself', '.', 'bias', 'ĊĠĠĠĠ']
Datasets feature(Train):
Dataset({
features: ['repo', 'path', 'func_name', 'original_string', 'language', 'code', 'code_tokens', 'docstring', 'docstring_tokens', 'sha', 'url', 'partition'],
num_rows: 412178
})