update dataset reading script.

NASA_OSDR.py

@@ -544,10 +544,8 @@ class NasaOsdr(datasets.GeneratorBasedBuilder):
             ]
 
     def _generate_examples(self, file: str):
-        idx = 0
-        for df_chunk in pd.read_csv(file, index_col=None, chunksize=1000):
-            for _, row in df_chunk.iterrows():
-                _item = {**row.to_dict()}
-                yield idx, _item
-                idx += 1
+        df = pd.read_csv(file, index_col=None)
+        for idx, row in df.iterrows():
+            _item = {**row.to_dict()}
+            yield idx, _item
 

tests/test_nasa_osdr.py

@@ -32,16 +32,15 @@ def test_inherit_generator_base_builder_works():
 
 
 def test_generate_examples():
-    dataset = NasaOsdr()
-    # train_data = dataset._generate_examples(assays_file='data/assays_all.csv', samples_file='data/samples.csv')
-    train_data = dataset._generate_examples(assays_file='data/train/samples.csv')
+    dataset = NasaOsdr(name='experiments')
+    train_data = dataset._generate_examples(file='data/train/experiments.csv')
     idx, sample = next(iter(train_data))
     print(idx, sample.keys())
 
 
 def test_load_dataset():
     from datasets import load_dataset
-    dataset = load_dataset("nasa_osdr.py")
+    dataset = load_dataset("nasa_osdr.py", "experiments")
     print(len(dataset["train"]))