Delete dataset_loader

dataset_loader

@@ -1,25 +0,0 @@
-from datasets import DatasetInfo, GeneratorBasedBuilder, SplitGenerator, Split, Features, Value
-
-class RomanshGerman(GeneratorBasedBuilder):
-    def _info(self):
-        return DatasetInfo(
-            description="Best Matched Parallel Data",
-            features=Features({
-                "romansh_text": Value("string"),
-                "german_text": Value("string"),
-                "similarity": Value("float32"),
-                "original_similarity": Value("float32"),
-            }),
-            supervised_keys=None,
-        )
-
-
-    def _split_generators(self, dl_manager):
-        data_path = dl_manager.download_and_extract("parallel_data/matched_0.60_0.005.jsonl")
-        return [SplitGenerator(name=Split.TRAIN, gen_kwargs={"filepath": data_path})]
-
-    def _generate_examples(self, filepath):
-        import json
-        with open(filepath, encoding="utf-8") as f:
-            for idx, line in enumerate(f):
-                yield idx, json.loads(line)