Update BubbleML_2.py

BubbleML_2.py

@@ -85,25 +85,27 @@ class BubbleMLDataset(GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager: datasets.DownloadManager):
         # data already lives in the repo
-        base_dir = dl_manager.extract("./")
+        base_dir = os.path.dirname(os.path.abspath(__file__))
+
         cfg = self.config
 
-        def _resolve(files_dict):
+        def resolve(files_dict):
             paths = []
-            for d, files in files_dict.items():
-                for fname in files:
-                    paths.append(os.path.join(base_dir, d, fname))
+            for subdir, flist in files_dict.items():
+                subdir_path = os.path.join(base_dir, subdir)
+                for fname in flist:
+                    full = os.path.join(subdir_path, fname)
+                    if not os.path.isfile(full):
+                        raise FileNotFoundError(f"Could not find {full}")
+                    paths.append(full)
             return paths
 
+        train_files = resolve(cfg.train_files_dict)
+        test_files  = resolve(cfg.test_files_dict)
+
         return [
-            datasets.SplitGenerator(
-                name=Split.TRAIN,
-                gen_kwargs={"files": _resolve(cfg.train_files_dict)},
-            ),
-            datasets.SplitGenerator(
-                name=Split.TEST,
-                gen_kwargs={"files": _resolve(cfg.test_files_dict)},
-            ),
+            datasets.SplitGenerator(name=Split.TRAIN, gen_kwargs={"files": train_files}),
+            datasets.SplitGenerator(name=Split.TEST,  gen_kwargs={"files": test_files}),
         ]
 
     def _generate_examples(self, files):