Convert dataset to Parquet (#4)

- Convert dataset to Parquet (6f0c473da12c0e552c0cbd8d7e1bb056f9efcf50)
- Delete loading script (c0c13575e32a4ca7350c1d1aab65d4e5f857ba60)

README.md

@@ -30,10 +30,15 @@ dataset_info:
           '1': positive
   splits:
   - name: train
-    num_bytes: 43369710
+    num_bytes: 43369614
     num_examples: 235165
-  download_size: 13184332
-  dataset_size: 43369710
+  download_size: 24354762
+  dataset_size: 43369614
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
 ---
 
 # Dataset Card for Turkish Product Reviews

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5ea59cd4f78f03895291c8a7f8be30b78b2f511c71569b68aef3a4f7756ee80
+size 24354762

turkish_product_reviews.py

@@ -1,60 +0,0 @@
-"""Turkish Product Reviews"""
-
-
-import os
-
-import datasets
-from datasets.tasks import TextClassification
-
-
-logger = datasets.logging.get_logger(__name__)
-
-
-_CITATION = ""
-
-_DESCRIPTION = """
-Turkish Product Reviews.
-This repository contains 235.165 product reviews collected online. There are 220.284 positive, 14881 negative reviews.
-"""
-
-_URL = "https://github.com/fthbrmnby/turkish-text-data/raw/master/reviews.tar.gz"
-_FILES_PATHS = ["reviews.pos", "reviews.neg"]
-
-_HOMEPAGE = "https://github.com/fthbrmnby/turkish-text-data"
-
-
-class TurkishProductReviews(datasets.GeneratorBasedBuilder):
-    VERSION = datasets.Version("1.0.0")
-
-    def _info(self):
-        return datasets.DatasetInfo(
-            description=_DESCRIPTION,
-            features=datasets.Features(
-                {
-                    "sentence": datasets.Value("string"),
-                    "sentiment": datasets.ClassLabel(names=["negative", "positive"]),
-                }
-            ),
-            citation=_CITATION,
-            homepage=_HOMEPAGE,
-            task_templates=[TextClassification(text_column="sentence", label_column="sentiment")],
-        )
-
-    def _split_generators(self, dl_manager):
-        """Returns SplitGenerators."""
-        archive = dl_manager.download(_URL)
-        return [
-            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"files": dl_manager.iter_archive(archive)}),
-        ]
-
-    def _generate_examples(self, files):
-        """Generate TurkishProductReviews examples."""
-        for file_idx, (path, f) in enumerate(files):
-            _, file_extension = os.path.splitext(path)
-            label = "negative" if file_extension == ".neg" else "positive"
-            for idx, line in enumerate(f):
-                line = line.decode("utf-8").strip()
-                yield f"{file_idx}_{idx}", {
-                    "sentence": line,
-                    "sentiment": label,
-                }