Pandas FutureWarning With ray.data.Dataset.map

Ray Libraries (Data, Train, Tune, Serve)
1

Hello,

I’m processing data as follows (my_func and fn_kwargs are just placeholders):

from ray import data

train_data = data.read_csv("file_name.csv")
train_data_mapped = train_data.map(my_func, fn_kwargs={"arg1":val1})
train_data_mapped.show(10)

I get the following depreciation warnings with Pandas versions >= 2.0.1 (it’s ok with <=2.0.3):

.../python3.9/site-packages/ray/data/_internal/pandas_block.py:67: FutureWarning: Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`
  if isinstance(items[0], TensorArrayElement):
.../python3.9/site-packages/ray/data/_internal/pandas_block.py:89: FutureWarning: Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`
  return items[0]