Delete stack_edu_huggingface.py

stack_edu_huggingface.py

@@ -1,104 +0,0 @@
-# stack_edu_huggingface.py
-
-import datasets
-
-_DESCRIPTION = """
-Pre-downloaded and processed version of the Stack-Edu dataset, which is a 125B token dataset of educational code filtered from The Stack v2.
-This dataset was curated by the Hugging Face team using a classifier-based filtering strategy to retain only the highest-quality educational programming content.
-This repository contains the raw text content organized by programming language in Parquet format, making it directly accessible without needing to re-download from Software Heritage.
-"""
-
-_CITATION = """
-@misc{allal2025smollm2smolgoesbig,
-      title={SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model},
-      author={...},
-      year={2025},
-      eprint={2502.02737},
-      archivePrefix={arXiv},
-      primaryClass={cs.CL}
-}
-"""
-
-_HOMEPAGE = "https://huggingface.co/datasets/HuggingFaceTB/stack-edu"
-_LICENSE = "ODC-By v1.0, also subject to CommonCrawl's Terms of Use and Software Heritage licenses."
-
-
-class StackEduConfig(datasets.BuilderConfig):
-    """BuilderConfig for Stack-Edu languages."""
-
-    def __init__(self, language, **kwargs):
-        """BuilderConfig for Stack-Edu.
-        Args:
-          language: string, the language to load
-          **kwargs: keyword arguments forwarded to super.
-        """
-        super(StackEduConfig, self).__init__(
-            name=language,
-            version=datasets.Version("1.0.0"),
-            **kwargs,
-        )
-        self.language = language
-
-
-class StackEdu(datasets.GeneratorBasedBuilder):
-    """Processed Stack-Edu Dataset"""
-
-    VERSION = datasets.Version("1.0.0")
-
-    # Define a configuration for each programming language
-    LANGUAGES = [
-        "c", "cpp", "csharp", "go", "java", "javascript", "markdown",
-        "php", "python", "ruby", "rust", "shell", "sql", "swift", "typescript"
-    ]
-
-    BUILDER_CONFIGS = [
-        StackEduConfig(
-            language=lang,
-            description=f"Educational code data for the {lang} language."
-        ) for lang in LANGUAGES
-    ]
-
-    def _info(self):
-        return datasets.DatasetInfo(
-            description=_DESCRIPTION,
-            features=datasets.Features(
-                {
-                    # Defina aqui as colunas do seu Parquet.
-                    # Baseado na sua print, parece ser algo assim:
-                    "language": datasets.Value("string"),
-                    "src_encoding": datasets.Value("string"),
-                    "length_bytes": datasets.Value("int64"),
-                    "score": datasets.Value("float64"),
-                    "int_score": datasets.Value("int64"),
-                    "detected_licenses": datasets.Sequence(datasets.Value("string")),
-                    "license_type": datasets.Value("string"),
-                    "text": datasets.Value("string"),
-                }
-            ),
-            homepage=_HOMEPAGE,
-            license=_LICENSE,
-            citation=_CITATION,
-        )
-
-    def _split_generators(self, dl_manager):
-        # O Hub baixa os arquivos para você. Aqui, especificamos o padrão de arquivo.
-        # Ele vai encontrar o arquivo parquet correspondente à configuração de linguagem.
-        file_path = f"{self.config.language}.parquet"
-        return [
-            datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                gen_kwargs={"filepath": file_path},
-            ),
-        ]
-
-    def _generate_examples(self, filepath):
-        """Yields examples."""
-        # A biblioteca `datasets` lida com a leitura de Parquet nativamente
-        # então não precisamos de um gerador complexo.
-        # Esta função pode ser mais elaborada se precisarmos de processamento extra.
-        # Por agora, a biblioteca cuidará disso.
-        # A biblioteca `datasets` usará o leitor de Parquet para este gerador.
-        # Esta função precisa existir, mas a mágica acontece por baixo dos panos.
-        # Apenas para cumprir a API.
-        # A mágica está em como o `datasets` lida com Parquet.
-        pass