fix(eda):type error for npartitions

SultanOrazbayev · jinglinpeng · commit 57db1ede3194 · 2022-06-09T16:50:19.000-07:00
Using `np.ceil` return a float, which triggers errors downstream. ```python In [5]: import numpy as np In [6]: np.ceil(1.0001) Out[6]: 2.0 ``` For a reproducible error example, see this question: https://stackoverflow.com/questions/72453608/dataprep-eda-typeerror-please-provide-npartitions-as-an-int-or-possibly-as-non.
diff --git a/dataprep/clean/utils.py b/dataprep/clean/utils.py
@@ -6,6 +6,7 @@
 import dask.dataframe as dd
 import numpy as np
 import pandas as pd
+from math import ceil
 
 
 NULL_VALUES = {
@@ -69,7 +70,7 @@ def to_dask(df: Union[pd.DataFrame, dd.DataFrame]) -> dd.DataFrame:
         return df
 
     df_size = df.memory_usage(deep=True).sum()
-    npartitions = np.ceil(df_size / 128 / 1024 / 1024)  # 128 MB partition size
+    npartitions = ceil(df_size / 128 / 1024 / 1024)  # 128 MB partition size
     return dd.from_pandas(df, npartitions=npartitions)
 
 
