fix(eda): change dtype 'string' to 'object'

Author: yuzhenmao

dataprep/eda/create_report/formatter.py

@@ -18,7 +18,14 @@
 from ..distribution.compute.overview import calc_stats
 from ..distribution.compute.univariate import cont_comps, nom_comps
 from ..distribution.render import format_cat_stats, format_num_stats, format_ov_stats
-from ..dtypes import CATEGORICAL_DTYPES, Continuous, Nominal, detect_dtype, is_dtype
+from ..dtypes import (
+    CATEGORICAL_DTYPES,
+    Continuous,
+    Nominal,
+    detect_dtype,
+    is_dtype,
+    string_dtype_to_object,
+)
 from ..intermediate import Intermediate
 from ..missing import render_missing
 from ..missing.compute.nullivariate import compute_missing_nullivariate
@@ -51,6 +58,7 @@ def format_report(
     # pylint: disable=too-many-locals,too-many-statements
     with ProgressBar(minimum=1, disable=not progress):
         df = to_dask(df)
+        df = string_dtype_to_object(df)
         if mode == "basic":
             comps = format_basic(df)
         # elif mode == "full":

dataprep/eda/distribution/compute/__init__.py

@@ -5,7 +5,7 @@
 import dask.dataframe as dd
 import pandas as pd
 
-from ...dtypes import DTypeDef
+from ...dtypes import DTypeDef, string_dtype_to_object
 from ...intermediate import Intermediate
 from ...utils import to_dask
 from .bivariate import compute_bivariate
@@ -93,9 +93,9 @@ def compute(
         dtype = {"a": Continuous(), "b": "nominal"}
         or dtype = Continuous() or dtype = "Continuous" or dtype = Continuous()
     """  # pylint: disable=too-many-locals
-
     df = to_dask(df)
     df.columns = df.columns.astype(str)
+    df = string_dtype_to_object(df)
 
     if not any((x, y, z)):
         return compute_overview(df, bins, ngroups, largest, timeunit, dtype)

dataprep/eda/dtypes.py

@@ -15,6 +15,9 @@
 CATEGORICAL_PANDAS_DTYPES = [pd.CategoricalDtype, pd.PeriodDtype]
 CATEGORICAL_DTYPES = CATEGORICAL_NUMPY_DTYPES + CATEGORICAL_PANDAS_DTYPES
 
+STRING_PANDAS_DTYPES = [pd.StringDtype]
+STRING_DTYPES = STRING_PANDAS_DTYPES
+
 NUMERICAL_NUMPY_DTYPES = [np.number]
 NUMERICAL_DTYPES = NUMERICAL_NUMPY_DTYPES
 
@@ -256,6 +259,17 @@ def is_pandas_categorical(dtype: Any) -> bool:
     return any(isinstance(dtype, c) for c in CATEGORICAL_PANDAS_DTYPES)
 
 
+def string_dtype_to_object(df: dd.DataFrame) -> dd.DataFrame:
+    """
+    Convert string dtype to object dtype
+    """
+    for col in df.columns:
+        if any(isinstance(df[col].dtype, c) for c in STRING_DTYPES):
+            df[col] = df[col].astype(object)
+
+    return df
+
+
 def drop_null(
     var: Union[dd.Series, pd.DataFrame, dd.DataFrame]
 ) -> Union[pd.Series, dd.Series, pd.DataFrame, dd.DataFrame]:

dataprep/eda/missing/compute/__init__.py

@@ -5,7 +5,7 @@
 from warnings import catch_warnings, filterwarnings
 
 from ...data_array import DataArray, DataFrame
-from ...dtypes import DTypeDef
+from ...dtypes import DTypeDef, string_dtype_to_object
 from ...intermediate import Intermediate
 from .bivariate import compute_missing_bivariate
 from .nullivariate import compute_missing_nullivariate
@@ -53,6 +53,7 @@ def compute_missing(  # pylint: disable=too-many-arguments
     >>> plot_missing(df, "HDI_for_year")
     >>> plot_missing(df, "HDI_for_year", "population")
     """
+    df = string_dtype_to_object(df)
     df = DataArray(df)
 
     # pylint: disable=no-else-raise