Delete dataset_loader.py

dataset_loader.py

@@ -1,78 +0,0 @@
-import json
-import datasets
-from datasets.download.download_manager import DownloadManager
-
-class AquaLLMDatasetConfig(datasets.BuilderConfig):
-    """AquaLLM Dataset configuration."""
-    def __init__(self, **kwargs):
-        super(AquaLLMDatasetConfig, self).__init__(version=datasets.Version("1.0.0"), **kwargs)
-
-class AquaLLMDataset(datasets.GeneratorBasedBuilder):
-    """AquaLLM Dataset generator."""
-    BUILDER_CONFIGS = [AquaLLMDatasetConfig(name="default", description="Default config for AquaLLM")]
-
-    def _info(self):
-        # Bu özellikler, _generate_examples'ın çıktısıyla TAM olarak eşleşmelidir.
-        return datasets.DatasetInfo(
-            features=datasets.Features({
-                "text": datasets.Value("string"),
-            }),
-            description="""
-            AquaLLM is a dataset for training a language model on various topics related to aquariums.
-            The dataset is structured with nested categories. This loader script flattens the nested JSON
-            and converts each entry into a single text string for training.
-            """,
-            homepage="https://huggingface.co/datasets/oytunistrator/AquaLLM-Dataset",
-            license="MIT",
-            citation="""
-            @misc{AquaLLMDataset,
-              title = {AquaLLM Dataset},
-              author = {Oytunistrator},
-              year = {2024},
-              howpublished = {Hugging Face Datasets Hub},
-              url = {https://huggingface.co/datasets/oytunistrator/AquaLLM-Dataset}
-            }
-            """
-        )
-
-    def _split_generators(self, dl_manager: DownloadManager):
-        """Returns SplitGenerators."""
-        data_file_path = dl_manager.download_and_extract(self.config.data_files["train"])
-        return [
-            datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                gen_kwargs={"data_file": data_file_path},
-            )
-        ]
-
-    def _generate_examples(self, data_file):
-        """Yields examples as dicts."""
-        with open(data_file, encoding="utf-8") as f:
-            data = json.load(f)
-
-        data_count = 0
-
-        # Helper function to create text from a dictionary
-        def create_text_from_dict(item_dict):
-            text = ""
-            for key, value in item_dict.items():
-                if isinstance(value, dict):
-                    # Handle nested dictionaries like "sicaklik_c"
-                    nested_text = ", ".join([f"{k}: {v}" for k, v in value.items()])
-                    text += f"{key.replace('_', ' ').capitalize()}: {nested_text}. "
-                elif isinstance(value, list):
-                    # Handle lists like "uyumlu_turler"
-                    list_text = ", ".join(value)
-                    text += f"{key.replace('_', ' ').capitalize()}: {list_text}. "
-                else:
-                    text += f"{key.replace('_', ' ').capitalize()}: {value}. "
-            return text.strip()
-
-        # Iterate through categories and their data
-        for category, item_list in data.items():
-            # Doğrudan listenin üzerinde döngüye girerek her bir öğeyi işle
-            if isinstance(item_list, list):
-                for item in item_list:
-                    text_data = create_text_from_dict(item)
-                    yield data_count, {"text": text_data}
-                    data_count += 1