metadata
annotations_creators: []
language_creators:
- crowdsourced
language:
- code
multilinguality:
- multilingual
size_categories:
- unknown
source_datasets: []
task_categories:
- text-generation
task_ids:
- language-modeling
Dataset Description
A subset of the-stack dataset, from 87 programming languages, and 295 extensions.
Each language is in a separate folder under data/ and contains folders of its extensions. We select samples from 20,000 random files of the original dataset, and keep a
maximum of 1,000 files per extension.
Check this space for inspecting this dataset.
Languages
The dataset contains 87 programming languages:
'ada', 'agda', 'alloy', 'antlr', 'applescript', 'assembly', 'augeas', 'awk', 'batchfile', 'bison', 'bluespec', 'c',
'c++', 'c-sharp', 'clojure', 'cmake', 'coffeescript', 'common-lisp', 'css', 'cuda', 'dart', 'dockerfile', 'elixir',
'elm', 'emacs-lisp','erlang', 'f-sharp', 'fortran', 'glsl', 'go', 'groovy', 'haskell','html', 'idris', 'isabelle', 'java',
'java-server-pages', 'javascript', 'julia', 'kotlin', 'lean', 'literate-agda', 'literate-coffeescript', 'literate-haskell',
'lua', 'makefile', 'maple', 'markdown', 'mathematica', 'matlab', 'ocaml', 'pascal', 'perl', 'php', 'powershell', 'prolog',
'protocol-buffer', 'python', 'r', 'racket', 'restructuredtext', 'rmarkdown', 'ruby', 'rust', 'sas', 'scala', 'scheme',
'shell', 'smalltalk', 'solidity', 'sparql', 'sql', 'stan', 'standard-ml', 'stata', 'systemverilog', 'tcl', 'tcsh', 'tex',
'thrift', 'typescript', 'verilog', 'vhdl', 'visual-basic', 'xslt', 'yacc', 'zig'
Dataset Structure
You can specify which language and extension you want to load:
# to load py extension of python
from datasets import load_dataset
load_dataset("bigcode/the-stack-inspection-data", data_dir="data/python/py")
DatasetDict({
train: Dataset({
features: ['content', 'lang', 'size', 'ext', 'max_stars_count', 'avg_line_length', 'max_line_length', 'alphanum_fraction'],
num_rows: 1000
})
})