switch np.where condition to check if non nan

Author: ryankarlos

pandas/core/reshape/tile.py

@@ -444,7 +444,7 @@ def _coerce_to_type(x):
     if dtype is not None:
         # GH 19768: force NaT to NaN during integer conversion
         if is_bool_dtype(x):
-            x = np.where(x.notna(), x.astype(int), np.nan)
+            x = np.where(~np.isnan(x), x.astype(int), np.nan)
         else:
             x = np.where(x.notna(), x.view(np.int64), np.nan)