fix(eda.data_array): handle empty df correctly

sfu-db · Sep 9, 2020 · 97db86d · 97db86d
1 parent a6b6e3b
commit 97db86d
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 6 deletions.
diff --git a/dataprep/eda/create_report/formatter.py b/dataprep/eda/create_report/formatter.py
@@ -2,6 +2,7 @@
 for create_report(df) function."""
 
 from typing import Any, Dict, List, Optional, Tuple, Union
+from warnings import catch_warnings, filterwarnings
 
 import dask
 import dask.dataframe as dd
@@ -80,7 +81,13 @@ def format_basic(df: dd.DataFrame) -> Dict[str, Any]:
     # aggregate all computations
     data, completions = basic_computations(df)
 
-    (data,) = dask.compute(data)
+    with catch_warnings():
+        filterwarnings(
+            "ignore",
+            "invalid value encountered in true_divide",
+            category=RuntimeWarning,
+        )
+        (data,) = dask.compute(data)
 
     # results dictionary
     res: Dict[str, Any] = {}

diff --git a/dataprep/eda/data_array.py b/dataprep/eda/data_array.py
@@ -196,7 +196,9 @@ def __init__(
 
         if isinstance(df, dd.DataFrame):
             is_pandas = False
-            self._ddf = df.astype({col: np.object for col in cat_cols})
+            if cat_cols and df.shape[1] != 0:
+                df = df.astype({col: np.object for col in cat_cols})
+            self._ddf = df
         elif isinstance(df, DataArray):
             self._ddf = df._ddf
             self._values = df._values
@@ -329,10 +331,11 @@ def __getitem__(self, indexer: Union[Sequence[str], str]) -> "DataArray":
         df = DataArray(subdf)
         df._values = self.values[:, cidx]  # pylint: disable=W0212
 
-        # coerce the array to it's minimal type
-        dtype = reduce(np.promote_types, df.dtypes.values)
-        if df._values.dtype != dtype:
-            df._values = df._values.astype(dtype)
+        if df.shape[1] != 0:
+            # coerce the array to it's minimal type
+            dtype = reduce(np.promote_types, df.dtypes.values)
+            if df._values.dtype != dtype:
+                df._values = df._values.astype(dtype)
 
         df._nulls = self.nulls[:, cidx]  # pylint: disable=W0212
         if self._head is not None: