WIP: preparing to sync with remote

view_dataset.py

@@ -0,0 +1,7 @@
+from datasets import load_dataset
+
+# Use the custom dataset script from the remote repo
+ds = load_dataset("shirve13/Demo", data_dir=".", download_mode="force_redownload", trust_remote_code=True)
+
+# Print a sample
+print(ds["train"][0])

visionreasoner_demo.py

@@ -1,36 +0,0 @@
-import os
-import datasets
-
-class VisionReasonerDataset(datasets.GeneratorBasedBuilder):
-    def _info(self):
-        return datasets.DatasetInfo(
-            features=datasets.Features({
-                "id": datasets.Value("string"),
-                "problem": datasets.Value("string"),
-                "solution": datasets.Value("string"),
-                "image": datasets.Image(),  # Enables image previews
-                "img_height": datasets.Value("int32"),
-            }),
-        )
-
-    def _split_generators(self, dl_manager):
-        parquet_path = os.path.join(dl_manager.manual_dir, "visionreasoner_dataset.parquet")
-        return [
-            datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                gen_kwargs={"filepath": parquet_path}
-            ),
-        ]
-
-    def _generate_examples(self, filepath):
-        import pandas as pd
-        df = pd.read_parquet(filepath)
-        for idx, row in df.iterrows():
-            yield idx, {
-                "id": row["id"],
-                "problem": row["problem"],
-                "solution": row["solution"],
-                "image": row["image"]["path"],
-                "img_height": row["img_height"],
-            }
-