[Batched Storage] Reset index after slicing dataframe in storage.py (#444)

* [Batched Storage] Reset index after slicing dataframe in storage.py

Fix indexing to reset index after slicing dataframe.

* add comments for reset_index

---------

Co-authored-by: Ma, Xiaochen <mxch1122@126.com>

Author: fatty-belly

dataflow/utils/storage.py

@@ -1018,7 +1018,7 @@ def read(self, output_type: Literal["dataframe", "dict"]="dataframe") -> Any:
         if self.batch_size:
             dataframe = dataframe.iloc[
                 self.batch_step * self.batch_size : (self.batch_step + 1) * self.batch_size
-            ]
+            ].reset_index(drop=True) # reset_index, make the new index of every batch start from 0. Instead of its original index.
         return self._convert_output(dataframe, output_type)
 
     def write(self, data: Any) -> Any:
@@ -1181,4 +1181,4 @@ def iter_chunks(self) -> Generator[pd.DataFrame, None, None]:
             for chunk in reader:
                 yield chunk
         else:
-            yield reader # 如果不支持流式，就产出整个 DF
+            yield reader # 如果不支持流式，就产出整个 DF