From 1f48f9f05246461f135d41dfb3a45c94dd2f762f Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Tue, 5 Mar 2024 16:44:02 +0800
Subject: [PATCH 01/19] add data processor logs

---
 sdgx/synthesizer.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/sdgx/synthesizer.py b/sdgx/synthesizer.py
index d17b34f5..61f10ba4 100644
--- a/sdgx/synthesizer.py
+++ b/sdgx/synthesizer.py
@@ -288,8 +288,10 @@ def fit(
         self.metadata = metadata  # Ensure update metadata
 
         logger.info("Fitting data processors...")
+        start_time = time.time()
         for d in self.data_processors:
             d.fit(metadata)
+        logger.info(f"Fitted {len(self.data_processors)} data processors in  {time.time() - start_time}s.")
 
         def chunk_generator() -> Generator[pd.DataFrame, None, None]:
             for chunk in self.dataloader.iter():

From 2eab25caf7ddec187bf266e2c4d5de078084285f Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Tue, 5 Mar 2024 16:44:20 +0800
Subject: [PATCH 02/19] intro NaN processor

---
 sdgx/data_processors/transformers/nan.py | 47 ++++++++++++++++++++++++
 1 file changed, 47 insertions(+)
 create mode 100644 sdgx/data_processors/transformers/nan.py

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
new file mode 100644
index 00000000..05ce98a9
--- /dev/null
+++ b/sdgx/data_processors/transformers/nan.py
@@ -0,0 +1,47 @@
+from __future__ import annotations
+
+from pandas import DataFrame  
+from sdgx.data_models.metadata import Metadata  
+from sdgx.data_processors.transformers.base import Transformer 
+from sdgx.data_processors.extension import hookimpl
+from sdgx.utils import logger
+
+class NonValueTransformer(Transformer):
+    '''
+    Transformer class for handling missing values in data.
+
+    This Transformer is mainly used as a reference for Transformer to facilitate developers to quickly understand the role of Transformer.
+    '''
+
+    def fit(self, metadata: Metadata | None = None):
+        '''
+        Fit method for the transformer. 
+        
+        Does not require any action.
+        '''
+        logger.info("Fitting NonValueTransformer")
+
+        return 
+
+    def convert(self, raw_data: DataFrame) -> DataFrame:
+        '''
+        Convert method to handle missing values in the input data.
+        '''
+        logger.info("Converting data using NonValueTransformer")
+
+        return raw_data.fillna()  
+    
+    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
+        '''
+        Reverse_convert method for the transformer. 
+        
+        Does not require any action.
+        '''
+        return processed_data
+
+    pass
+
+@hookimpl
+def register(manager):
+    manager.register("NonValueTransformer", NonValueTransformer)
+

From 6712ae91ee3f2be154d7d9eb90164a6a6bd8a1c5 Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 11:48:49 +0800
Subject: [PATCH 03/19] fix fill_na typo in NanValueTransformer

---
 sdgx/data_processors/transformers/nan.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index 05ce98a9..731c8ee7 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -13,6 +13,8 @@ class NonValueTransformer(Transformer):
     This Transformer is mainly used as a reference for Transformer to facilitate developers to quickly understand the role of Transformer.
     '''
 
+    fill_na_value = 0 
+
     def fit(self, metadata: Metadata | None = None):
         '''
         Fit method for the transformer. 
@@ -29,7 +31,7 @@ def convert(self, raw_data: DataFrame) -> DataFrame:
         '''
         logger.info("Converting data using NonValueTransformer")
 
-        return raw_data.fillna()  
+        return raw_data.fillna(value= self.fill_na_value)  
     
     def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
         '''

From f16d12ba1a17ae2541e884d8714468823cba30bb Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 11:49:16 +0800
Subject: [PATCH 04/19] add default_processors in manager

---
 sdgx/data_processors/manager.py | 63 ++++++++++++++++++++++++++++++++-
 1 file changed, 62 insertions(+), 1 deletion(-)

diff --git a/sdgx/data_processors/manager.py b/sdgx/data_processors/manager.py
index 3a2e8f8f..50df94ef 100644
--- a/sdgx/data_processors/manager.py
+++ b/sdgx/data_processors/manager.py
@@ -8,21 +8,82 @@
 from sdgx.data_processors.extension import project_name as PROJECT_NAME
 from sdgx.manager import Manager
 
-
 class DataProcessorManager(Manager):
+    """
+    This is a plugin management class for data processing components.
+    
+    Properties: 
+        - register_type: Specifies the type of data processors to register.
+        - project_name: Stores the project name from the extension module.
+        - hookspecs_model: Stores the hook specifications model from the extension module.
+        - preset_default_processors: Stores a list of default processor names in lowercase.
+        - registed_data_processors: Property that returns the registered data processors.
+        - registed_default_processor_list: Property that returns the registered default data processors.
+
+    Methods:
+        - load_all_local_model: Loads all local models for formatters, generators, samplers, and transformers.
+        - init_data_processor: Initializes a data processor with the given name and keyword arguments.
+        - init_all_processors: Initializes all registered data processors with optional keyword arguments.
+        - init_default_processors: Initializes default processors that are both registered and preset.
+
+    """
     register_type = DataProcessor
+    
     project_name = PROJECT_NAME
+    
     hookspecs_model = extension
 
+    preset_defalut_processors = [ p.lower() for p in ["NonValueTransformer"]]
+    
     @property
     def registed_data_processors(self):
+        '''
+        This property returns all registered data processors
+        '''
         return self.registed_cls
+    
+    @property
+    def registed_default_processor_list(self):
+        '''
+        This property returns all registered default data processors
+        '''
+        registed_processor_list = self.registed_data_processors.keys()
+        
+        # calculate intersection 
+        target_processors = list(set(registed_processor_list).intersection(self.preset_defalut_processors))
+        
+        return target_processors
 
     def load_all_local_model(self):
+        '''
+        loads all local models
+        '''
         self._load_dir(data_processors.formatters)
         self._load_dir(data_processors.generators)
         self._load_dir(data_processors.samplers)
         self._load_dir(data_processors.transformers)
 
     def init_data_processor(self, processor_name, **kwargs: dict[str, Any]) -> DataProcessor:
+        '''
+        Initializes a data processor with the given name and parameters
+        '''
         return self.init(processor_name, **kwargs)
+
+    def init_all_processors(self, **kwargs: Any) -> list[DataProcessor]:
+        '''
+        Initializes all registered data processors
+        '''
+        return [
+            self.init(processor_name, **kwargs)
+            for processor_name in self.registed_data_processors.keys()
+        ]
+
+    def init_default_processors(self, **kwargs: Any) -> list[DataProcessor]:
+        '''
+        Initializes all default data processors
+        '''
+
+        return [
+            self.init(processor_name, **kwargs)
+            for processor_name in self.registed_default_processor_list
+        ]
\ No newline at end of file

From 4510d8572b35f2d396e0b418e69bdc3735a7496c Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 11:49:28 +0800
Subject: [PATCH 05/19] add default_processors in synthesizer

---
 sdgx/synthesizer.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sdgx/synthesizer.py b/sdgx/synthesizer.py
index 61f10ba4..0f38c72b 100644
--- a/sdgx/synthesizer.py
+++ b/sdgx/synthesizer.py
@@ -99,9 +99,9 @@ def __init__(
             self.dataloader = None
 
         # Init data processors
-        if not data_processors:
-            data_processors = []
         self.data_processors_manager = DataProcessorManager()
+        if not data_processors:
+            data_processors = self.data_processors_manager.registed_default_processor_list
         self.data_processors = [
             (
                 d

From 9afe27a5fc13522865019550198d928b28cca4ee Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 11:52:25 +0800
Subject: [PATCH 06/19] update log in nan

---
 sdgx/data_processors/transformers/nan.py | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index 731c8ee7..f75c8c7f 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -21,7 +21,7 @@ def fit(self, metadata: Metadata | None = None):
         
         Does not require any action.
         '''
-        logger.info("Fitting NonValueTransformer")
+        logger.info("NonValueTransformer Fitted.")
 
         return 
 
@@ -29,9 +29,14 @@ def convert(self, raw_data: DataFrame) -> DataFrame:
         '''
         Convert method to handle missing values in the input data.
         '''
-        logger.info("Converting data using NonValueTransformer")
 
-        return raw_data.fillna(value= self.fill_na_value)  
+        logger.info("Converting data using NonValueTransformer...")
+
+        res = raw_data.fillna(value= self.fill_na_value)  
+        
+        logger.info("Converting data using NonValueTransformer... Finished.")
+
+        return res
     
     def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
         '''
@@ -39,6 +44,8 @@ def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
         
         Does not require any action.
         '''
+        logger.info("Data reverse-converted by NonValueTransformer (No Action).")
+
         return processed_data
 
     pass

From da08ff1b7b04c1fb0442b0d3193518b6d6f56e7c Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 15:23:03 +0800
Subject: [PATCH 07/19] update formatter base

---
 sdgx/data_processors/formatters/base.py | 28 +++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/sdgx/data_processors/formatters/base.py b/sdgx/data_processors/formatters/base.py
index 52cdf320..201df19d 100644
--- a/sdgx/data_processors/formatters/base.py
+++ b/sdgx/data_processors/formatters/base.py
@@ -1,3 +1,6 @@
+from __future__ import annotations
+
+import pandas as pd 
 from sdgx.data_processors.base import DataProcessor
 
 
@@ -16,3 +19,28 @@ class Formatter(DataProcessor):
     - :ref:`Transformer` sometimes implements some functions with the help of Formatter.
 
     """
+
+    def reverse_convert(self, processed_data: pd.DataFrame) -> pd.DataFrame:
+        """Convert processed data into raw data.
+
+        Args:
+            processed_data (pd.DataFrame): Processed data
+
+        Returns:
+            pd.DataFrame: Raw data
+        """
+        return self.post_processing(processed_data)
+    
+     
+    def post_processing(self, processed_data: pd.DataFrame) -> pd.DataFrame:
+        '''
+        For formatter, please rewrite this method.
+        
+        Args:
+            processed_data (pd.DataFrame): Processed data
+
+        Returns:
+            pd.DataFrame: Raw data
+        '''
+
+        return processed_data

From e5ea5ab6bcf0d8f58ebbdf76b04adfaa39d50aed Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 15:23:13 +0800
Subject: [PATCH 08/19] add int formatter

---
 sdgx/data_processors/formatters/int.py | 58 ++++++++++++++++++++++++++
 sdgx/data_processors/manager.py        |  2 +-
 2 files changed, 59 insertions(+), 1 deletion(-)
 create mode 100644 sdgx/data_processors/formatters/int.py

diff --git a/sdgx/data_processors/formatters/int.py b/sdgx/data_processors/formatters/int.py
new file mode 100644
index 00000000..62f19645
--- /dev/null
+++ b/sdgx/data_processors/formatters/int.py
@@ -0,0 +1,58 @@
+from __future__ import annotations
+
+import pandas as pd 
+from sdgx.data_models.metadata import Metadata  
+from sdgx.data_processors.formatters.base import Formatter
+from sdgx.data_processors.extension import hookimpl
+from sdgx.utils import logger
+
+class IntValueFormatter(Formatter):
+    '''
+    Formatter class for handling Int values in pd.DataFrame.
+    '''
+
+    int_columns = None
+
+    def fit(self, metadata: Metadata | None = None):
+        '''
+        Fit method for the formatter. 
+        
+        Formatter need to use metadata to record which columns belong to the int type, and convert them back to the int type during post-processing.
+        '''
+        
+        # get from metadata 
+        self.int_columns = metadata.get("int_columns")
+        
+        logger.info("IntValueFormatter Fitted.")
+
+        return 
+
+    def convert(self, raw_data: pd.DataFrame) -> pd.DataFrame:
+        '''
+        Convert method to handle missing values in the input data.
+        '''
+        
+        logger.info("Converting data using IntValueFormatter... Finished  (No Action).")
+
+        return raw_data
+    
+    def post_processing(self, processed_data: pd.DataFrame) -> pd.DataFrame:
+        '''
+        post_processing method for the formatter. 
+        
+        Does not require any action.
+        '''
+
+        for col in self.int_columns:
+            processed_data[col] = processed_data[col].astype(int)
+
+        logger.info("Data reverse-converted by IntValueFormatter.")
+
+        return processed_data
+
+    pass
+
+@hookimpl
+def register(manager):
+    manager.register("IntValueFormatter", IntValueFormatter)
+
diff --git a/sdgx/data_processors/manager.py b/sdgx/data_processors/manager.py
index 50df94ef..4c75cdfc 100644
--- a/sdgx/data_processors/manager.py
+++ b/sdgx/data_processors/manager.py
@@ -33,7 +33,7 @@ class DataProcessorManager(Manager):
     
     hookspecs_model = extension
 
-    preset_defalut_processors = [ p.lower() for p in ["NonValueTransformer"]]
+    preset_defalut_processors = [ p.lower() for p in ["NonValueTransformer",'IntValueFormatter']]
     
     @property
     def registed_data_processors(self):

From 6b25bc1fe9227e31b97cf718ca758d1e11a69ba2 Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 16:44:52 +0800
Subject: [PATCH 09/19] Update nan.py

---
 sdgx/data_processors/transformers/nan.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index f75c8c7f..8f52c4bb 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -14,6 +14,8 @@ class NonValueTransformer(Transformer):
     '''
 
     fill_na_value = 0 
+    
+    drop_na = False
 
     def fit(self, metadata: Metadata | None = None):
         '''
@@ -32,7 +34,10 @@ def convert(self, raw_data: DataFrame) -> DataFrame:
 
         logger.info("Converting data using NonValueTransformer...")
 
-        res = raw_data.fillna(value= self.fill_na_value)  
+        if self.drop:
+            res = raw_data.dropna()
+        else:
+            res = raw_data.fillna(value= self.fill_na_value)  
         
         logger.info("Converting data using NonValueTransformer... Finished.")
 

From c9f70248eccffe8a1b8978ae09ac1055fa73e3dd Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Wed, 6 Mar 2024 18:19:49 +0800
Subject: [PATCH 10/19] Create discrete.py

still draft
---
 sdgx/data_processors/transformers/discrete.py | 61 +++++++++++++++++++
 1 file changed, 61 insertions(+)
 create mode 100644 sdgx/data_processors/transformers/discrete.py

diff --git a/sdgx/data_processors/transformers/discrete.py b/sdgx/data_processors/transformers/discrete.py
new file mode 100644
index 00000000..d91e14e0
--- /dev/null
+++ b/sdgx/data_processors/transformers/discrete.py
@@ -0,0 +1,61 @@
+from __future__ import annotations
+
+from pandas import DataFrame  
+from sdgx.data_models.metadata import Metadata  
+from sdgx.data_processors.transformers.base import Transformer 
+from sdgx.data_processors.extension import hookimpl
+from sdgx.utils import logger
+
+class DiscreteTransformer(Transformer):
+    '''
+    DiscreteTransformer is an important component of sdgx, used to handle discrete columns.
+    
+    By default, DiscreteTransformer will perform one-hot encoding of discrete columns, and issue a warning message when dimensionality explosion occurs.
+    '''
+
+    discrete_columns = None
+    '''
+    Record which columns are of discrete type.
+    '''
+
+    def fit(self, metadata: Metadata | None = None):
+        '''
+        Fit method for the transformer. 
+        '''
+
+        self.discrete_columns = metadata.get('discrete_columns')
+
+        logger.info("DiscreteTransformer Fitted.")
+        return 
+
+    def convert(self, raw_data: DataFrame) -> DataFrame:
+        '''
+        Convert method to handle discrete values in the input data.
+        '''
+
+        logger.info("Converting data using DiscreteTransformer...")
+
+        if self.drop:
+            res = raw_data.dropna()
+        else:
+            res = raw_data.fillna(value= self.fill_na_value)  
+        
+        logger.info("Converting data using DiscreteTransformer... Finished.")
+
+        return res
+    
+    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
+        '''
+        Reverse_convert method for the transformer. 
+        
+        
+        '''
+
+        logger.info("Data reverse-converted by DiscreteTransformer.")
+
+        return processed_data
+
+
+
+
+    pass
\ No newline at end of file

From 57b5a3e13e6f8e0f249da4aaecaa6017b68ca8ab Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Fri, 8 Mar 2024 11:50:44 +0800
Subject: [PATCH 11/19] update method typing

---
 sdgx/data_processors/formatters/int.py    |  4 +++-
 sdgx/data_processors/transformers/base.py | 14 ++++++++++++++
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/sdgx/data_processors/formatters/int.py b/sdgx/data_processors/formatters/int.py
index 62f19645..0ed4d35d 100644
--- a/sdgx/data_processors/formatters/int.py
+++ b/sdgx/data_processors/formatters/int.py
@@ -1,6 +1,8 @@
 from __future__ import annotations
 
 import pandas as pd 
+from typing import Any
+
 from sdgx.data_models.metadata import Metadata  
 from sdgx.data_processors.formatters.base import Formatter
 from sdgx.data_processors.extension import hookimpl
@@ -13,7 +15,7 @@ class IntValueFormatter(Formatter):
 
     int_columns = None
 
-    def fit(self, metadata: Metadata | None = None):
+    def fit(self, metadata: Metadata | None = None, **kwargs: dict[str, Any]):
         '''
         Fit method for the formatter. 
         
diff --git a/sdgx/data_processors/transformers/base.py b/sdgx/data_processors/transformers/base.py
index 0f4c1f96..cc39a47c 100644
--- a/sdgx/data_processors/transformers/base.py
+++ b/sdgx/data_processors/transformers/base.py
@@ -1,4 +1,8 @@
+import pandas as pd 
 from sdgx.data_processors.base import DataProcessor
+from sdgx.data_models.metadata import Metadata
+from sdgx.data_loader import DataLoader
+from sdgx.models.components.optimize.ndarray_loader import NDArrayLoader
 
 
 class Transformer(DataProcessor):
@@ -10,3 +14,13 @@ class Transformer(DataProcessor):
 
     To achieve that, Transformer can use :ref:`Formatter` and :ref:`Inspector` to help.
     """
+
+    def fit(self, metadata: Metadata | None = None, tabular_data: DataLoader | pd.DataFrame = None):
+        '''
+        Fit method for the transformer. 
+        
+        由于 ，Transformer 可能需要更多参数。
+        '''
+        
+
+        return  
\ No newline at end of file

From e12f1af203017f5a403e39b094ae2a5ccee7a37b Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Fri, 8 Mar 2024 13:12:40 +0800
Subject: [PATCH 12/19] Update discrete.py

---
 sdgx/data_processors/transformers/discrete.py | 70 ++++++++++++++++---
 1 file changed, 59 insertions(+), 11 deletions(-)

diff --git a/sdgx/data_processors/transformers/discrete.py b/sdgx/data_processors/transformers/discrete.py
index d91e14e0..1d6cbfe0 100644
--- a/sdgx/data_processors/transformers/discrete.py
+++ b/sdgx/data_processors/transformers/discrete.py
@@ -1,10 +1,14 @@
 from __future__ import annotations
 
-from pandas import DataFrame  
+import pandas as pd 
+from sklearn.preprocessing import LabelEncoder, OneHotEncoder
+
 from sdgx.data_models.metadata import Metadata  
 from sdgx.data_processors.transformers.base import Transformer 
 from sdgx.data_processors.extension import hookimpl
 from sdgx.utils import logger
+from sdgx.data_loader import DataLoader
+from sdgx.models.components.optimize.ndarray_loader import NDArrayLoader
 
 class DiscreteTransformer(Transformer):
     '''
@@ -13,38 +17,82 @@ class DiscreteTransformer(Transformer):
     By default, DiscreteTransformer will perform one-hot encoding of discrete columns, and issue a warning message when dimensionality explosion occurs.
     '''
 
-    discrete_columns = None
+    discrete_columns: list = None
     '''
     Record which columns are of discrete type.
     '''
 
-    def fit(self, metadata: Metadata | None = None):
+    encoders: dict = {}
+
+    onehot_encoder_handle_unknown='ignore'
+
+    def fit(self, metadata: Metadata, tabular_data: DataLoader | pd.DataFrame):
         '''
         Fit method for the transformer. 
         '''
 
+        logger.info("Fitting using DiscreteTransformer...")
+
         self.discrete_columns = metadata.get('discrete_columns')
+        
+        # no discrete columns 
+        if len(self.discrete_columns) == 0 :
+            logger.info("Fitting using DiscreteTransformer... Finished (No Columns).")
+            return 
+
+        # then, there are >= 1 discrete colums
+        for each_col in self.discrete_columns:
+            # fit each column 
+            self._fit_column(tabular_data[[each_col]])
 
-        logger.info("DiscreteTransformer Fitted.")
+        logger.info("Fitting using DiscreteTransformer... Finished.")
+        
         return 
+    
+    def _fit_column(self, column_name: str, column_data: pd.DataFrame):
+        '''
+        Fit every discrete columns in `_fit_column`.
+
+        Args:
+            - column_data (pd.DataFrame): A dataframe containing a column.
+            - column_name: str: column name.
 
-    def convert(self, raw_data: DataFrame) -> DataFrame:
+        '''
+
+        self.encoders[column_name] = OneHotEncoder(handle_unknown= self.onehot_encoder_handle_unknown)
+        # fit the column data
+        self.encoders[column_name].fit(column_data)
+        
+        logger.info(f"Discrete column {column_name} fitted.")
+        
+
+    def convert(self, raw_data: pd.DataFrame) -> pd.DataFrame:
         '''
         Convert method to handle discrete values in the input data.
         '''
 
         logger.info("Converting data using DiscreteTransformer...")
 
-        if self.drop:
-            res = raw_data.dropna()
-        else:
-            res = raw_data.fillna(value= self.fill_na_value)  
+        # TODO 
+        # transform every discrete column into 
         
         logger.info("Converting data using DiscreteTransformer... Finished.")
+        
+        # return the result
+        return 
+    
+    def _transform_column(self, column_name: str, column_data: pd.DataFrame | pd.Series):
+        '''
+        Transform every single discrete columns in `_transform_column`.
 
-        return res
+        Args:
+            - column_data (pd.DataFrame): A dataframe containing a column.
+            - column_name: str: column name.
+
+        '''
+        pass
     
-    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
+    def reverse_convert(self, processed_data: pd.DataFrame) -> pd.DataFrame:
         '''
         Reverse_convert method for the transformer. 
         

From 251f782b8f63f629a5d0b6ff7a65f2c514f265b9 Mon Sep 17 00:00:00 2001
From: MoooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 10:44:55 +0800
Subject: [PATCH 13/19] update fit method kargs

---
 sdgx/data_processors/transformers/nan.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index 8f52c4bb..53a3f012 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -1,6 +1,8 @@
 from __future__ import annotations
 
 from pandas import DataFrame  
+from typing import Any 
+
 from sdgx.data_models.metadata import Metadata  
 from sdgx.data_processors.transformers.base import Transformer 
 from sdgx.data_processors.extension import hookimpl
@@ -17,7 +19,7 @@ class NonValueTransformer(Transformer):
     
     drop_na = False
 
-    def fit(self, metadata: Metadata | None = None):
+    def fit(self, metadata: Metadata | None = None,  **kwargs: dict[str, Any]):
         '''
         Fit method for the transformer. 
         

From 52244b78432ddcfeb48c0cead4e69c733ce7e8e0 Mon Sep 17 00:00:00 2001
From: MoooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 10:49:44 +0800
Subject: [PATCH 14/19] fix typo in nan.py

---
 sdgx/data_processors/transformers/nan.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index 53a3f012..ff07f8c1 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -36,7 +36,7 @@ def convert(self, raw_data: DataFrame) -> DataFrame:
 
         logger.info("Converting data using NonValueTransformer...")
 
-        if self.drop:
+        if self.drop_na:
             res = raw_data.dropna()
         else:
             res = raw_data.fillna(value= self.fill_na_value)  

From e1f5d737073fd8920c85f7fe579bcaa2e5290db7 Mon Sep 17 00:00:00 2001
From: MoooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 10:50:16 +0800
Subject: [PATCH 15/19] intro col order, update base transformer

---
 sdgx/data_processors/transformers/base.py     | 14 ++++-
 .../transformers/column_order.py              | 58 +++++++++++++++++++
 2 files changed, 69 insertions(+), 3 deletions(-)
 create mode 100644 sdgx/data_processors/transformers/column_order.py

diff --git a/sdgx/data_processors/transformers/base.py b/sdgx/data_processors/transformers/base.py
index cc39a47c..cd1860e2 100644
--- a/sdgx/data_processors/transformers/base.py
+++ b/sdgx/data_processors/transformers/base.py
@@ -18,9 +18,17 @@ class Transformer(DataProcessor):
     def fit(self, metadata: Metadata | None = None, tabular_data: DataLoader | pd.DataFrame = None):
         '''
         Fit method for the transformer. 
-        
-        由于 ，Transformer 可能需要更多参数。
         '''
         
 
-        return  
\ No newline at end of file
+        return  
+
+    @staticmethod
+    def delete_column(tabular_data,  column_name):
+
+        pass
+
+    @staticmethod
+    def attach_columns(tabular_data, new_columns):
+
+        pass
\ No newline at end of file
diff --git a/sdgx/data_processors/transformers/column_order.py b/sdgx/data_processors/transformers/column_order.py
new file mode 100644
index 00000000..82e6fbdb
--- /dev/null
+++ b/sdgx/data_processors/transformers/column_order.py
@@ -0,0 +1,58 @@
+from __future__ import annotations
+
+from pandas import DataFrame  
+from sdgx.data_models.metadata import Metadata  
+from sdgx.data_processors.transformers.base import Transformer 
+from sdgx.data_processors.extension import hookimpl
+from sdgx.utils import logger
+
+class ColumnOrderTransformer(Transformer):
+    '''
+    Transformer class for handling missing values in data.
+
+    This Transformer is mainly used as a reference for Transformer to facilitate developers to quickly understand the role of Transformer.
+    '''
+
+    column_list: list = None
+    '''
+    The list of tabular data's columns.
+    '''
+
+    def fit(self, metadata: Metadata | None = None):
+        '''
+        Fit method for the transformer. 
+        
+        Remember the order of the columns.
+        '''
+
+        self.column_list = list(metadata.column_list)
+
+        logger.info("ColumnOrderTransformer Fitted.")
+
+        return 
+
+    def convert(self, raw_data: DataFrame) -> DataFrame:
+        '''
+        Convert method to handle missing values in the input data.
+        '''
+        logger.info("Converting data using ColumnOrderTransformer...")
+        logger.info("Converting data using ColumnOrderTransformer... Finished (No action).")
+
+        return raw_data
+    
+    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
+        '''
+        Reverse_convert method for the transformer. 
+        '''
+
+
+        logger.info("Data reverse-converted by ColumnOrderTransformer.")
+
+        return processed_data
+
+    pass
+
+@hookimpl
+def register(manager):
+    manager.register("ColumnOrderTransformer", ColumnOrderTransformer)
+

From 655e60f134b413527a73565c7ac67a632791e1dd Mon Sep 17 00:00:00 2001
From: MoooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 10:50:24 +0800
Subject: [PATCH 16/19] Update discrete.py

---
 sdgx/data_processors/transformers/discrete.py | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/sdgx/data_processors/transformers/discrete.py b/sdgx/data_processors/transformers/discrete.py
index 1d6cbfe0..b5e8e617 100644
--- a/sdgx/data_processors/transformers/discrete.py
+++ b/sdgx/data_processors/transformers/discrete.py
@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import pandas as pd 
-from sklearn.preprocessing import LabelEncoder, OneHotEncoder
+from sklearn.preprocessing import OneHotEncoder
 
 from sdgx.data_models.metadata import Metadata  
 from sdgx.data_processors.transformers.base import Transformer 
@@ -28,7 +28,7 @@ class DiscreteTransformer(Transformer):
 
     def fit(self, metadata: Metadata, tabular_data: DataLoader | pd.DataFrame):
         '''
-        Fit method for the transformer. 
+        Fit method for the DiscreteTransformer. 
         '''
 
         logger.info("Fitting using DiscreteTransformer...")
@@ -43,7 +43,7 @@ def fit(self, metadata: Metadata, tabular_data: DataLoader | pd.DataFrame):
         # then, there are >= 1 discrete colums
         for each_col in self.discrete_columns:
             # fit each column 
-            self._fit_column(tabular_data[[each_col]])
+            self._fit_column(each_col, tabular_data[[each_col]])
 
         logger.info("Fitting using DiscreteTransformer... Finished.")
         
@@ -56,7 +56,6 @@ def _fit_column(self, column_name: str, column_data: pd.DataFrame):
         Args:
             - column_data (pd.DataFrame): A dataframe containing a column.
             - column_name: str: column name.
-
         '''
 
         self.encoders[column_name] = OneHotEncoder(handle_unknown= self.onehot_encoder_handle_unknown)
@@ -75,6 +74,15 @@ def convert(self, raw_data: pd.DataFrame) -> pd.DataFrame:
 
         # TODO 
         # transform every discrete column into 
+        if len(self.discrete_columns) == 0:
+            logger.info("Converting data using DiscreteTransformer... Finished (No column).")
+            return 
+        
+        for each_col in self.discrete_columns:
+            new_onehot_column_set = self.encoders[each_col].transform(raw_data[[each_col]])
+            # TODO 1- add new_onehot_column_set into the original dataframe
+            # TODO 2- delete the original column 
+            logger.info(f"Column {each_col} converted.")
         
         logger.info("Converting data using DiscreteTransformer... Finished.")
         
@@ -103,7 +111,4 @@ def reverse_convert(self, processed_data: pd.DataFrame) -> pd.DataFrame:
 
         return processed_data
 
-
-
-
     pass
\ No newline at end of file

From 57237d1d50d79757d92fdaa8d00a8a5fa09c1a0a Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 11:55:14 +0800
Subject: [PATCH 17/19] intro ChinaMainlandAddressInspector

---
 sdgx/data_models/inspectors/personal.py | 27 +++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/sdgx/data_models/inspectors/personal.py b/sdgx/data_models/inspectors/personal.py
index fb8d2454..30b1ef97 100644
--- a/sdgx/data_models/inspectors/personal.py
+++ b/sdgx/data_models/inspectors/personal.py
@@ -73,6 +73,30 @@ def domain_verification(self, each_sample):
             return False
         return True
 
+class ChinaMainlandAddressInspector(RegexInspector):
+
+    # This regular expression does not take effect and is only for reference by developers. 
+    # pattern = r"^[\u4e00-\u9fa5]{2,}(省|自治区|特别行政区|市)|[\u4e00-\u9fa5]{2,}(市|区|县|自治州|自治县|县级市|地区|盟|林区)?|[\u4e00-\u9fa5]{0,}(街道|镇|乡)?|[\u4e00-\u9fa5]{0,}(路|街|巷|弄)?|[\u4e00-\u9fa5]{0,}(号|弄)?$"
+
+    pattern = r"^[\u4e00-\u9fa5]{2,}(省|自治区|特别行政区|市|县|村|弄|乡|路|街)"
+    
+    pii = True
+
+    data_type_name = "china_mainland_address"
+
+    _inspect_level = 30
+
+    def domain_verification(self, each_sample):
+        # CHN address should be between 8 - 30 characters
+        if len(each_sample) < 8:
+            return False
+        if len(each_sample) > 30:
+            return False
+        # notice to distinguishing from the company name
+        if each_sample.endswith('公司'):
+            return False
+        return True
+    
 
 @hookimpl
 def register(manager):
@@ -85,3 +109,6 @@ def register(manager):
     manager.register("ChinaMainlandPostCode", ChinaMainlandPostCode)
 
     manager.register("ChinaMainlandUnifiedSocialCreditCode", ChinaMainlandUnifiedSocialCreditCode)
+
+    manager.register("ChinaMainlandAddressInspector", ChinaMainlandAddressInspector)
+

From 90510d93851fdf377bd593f9f9fb7c4337c9397d Mon Sep 17 00:00:00 2001
From: MooooCat <gyz@guoyunzhe.com>
Date: Mon, 11 Mar 2024 11:55:22 +0800
Subject: [PATCH 18/19] add test cases

---
 tests/data_models/inspector/test_personal.py | 29 ++++++++++++++++----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/tests/data_models/inspector/test_personal.py b/tests/data_models/inspector/test_personal.py
index c7d27490..1477c87d 100644
--- a/tests/data_models/inspector/test_personal.py
+++ b/tests/data_models/inspector/test_personal.py
@@ -1,21 +1,21 @@
+import pytest
 import datetime
 import random
 import string
-
 import pandas as pd
-import pytest
+
 from faker import Faker
+fake = Faker(locale="zh_CN")
 
 from sdgx.data_models.inspectors.personal import (
+    EmailInspector,
     ChinaMainlandIDInspector,
     ChinaMainlandMobilePhoneInspector,
     ChinaMainlandPostCode,
     ChinaMainlandUnifiedSocialCreditCode,
-    EmailInspector,
+    ChinaMainlandAddressInspector
 )
 
-fake = Faker(locale="zh_CN")
-
 
 def generate_uniform_credit_code():
     # generate china mainland 统一社会信用代码 for test
@@ -193,6 +193,25 @@ def test_chn_uscc_inspector_generated_data(chn_personal_test_df: pd.DataFrame):
     assert inspector_USCC.inspect_level == 30
     assert inspector_USCC.pii is True
 
+# CHN address
+def test_chn_address_inspector_demo_data(raw_data):
+    inspector_CHN_Address = ChinaMainlandAddressInspector()
+    inspector_CHN_Address.fit(raw_data)
+    assert not inspector_CHN_Address.regex_columns
+    assert sorted(inspector_CHN_Address.inspect()["china_mainland_address_columns"]) == sorted([])
+    assert inspector_CHN_Address.inspect_level == 30
+    assert inspector_CHN_Address.pii is True
+
+
+def test_chn_address_inspector_generated_data(chn_personal_test_df: pd.DataFrame):
+    inspector_CHN_Address = ChinaMainlandAddressInspector()
+    inspector_CHN_Address.fit(chn_personal_test_df)
+    # assert inspector_CHN_Address.regex_columns
+    assert sorted(inspector_CHN_Address.inspect()["china_mainland_address_columns"]) == sorted(
+        ["chn_address"]
+    )
+    assert inspector_CHN_Address.inspect_level == 30
+    assert inspector_CHN_Address.pii is True
 
 if __name__ == "__main__":
     pytest.main(["-vv", "-s", __file__])

From 4af24aba8ed3c525fbee5686c80dab00dadb8b77 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Mon, 11 Mar 2024 03:56:41 +0000
Subject: [PATCH 19/19] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 sdgx/data_models/inspectors/personal.py       | 10 +-
 sdgx/data_processors/formatters/base.py       | 12 +--
 sdgx/data_processors/formatters/int.py        | 46 ++++-----
 sdgx/data_processors/manager.py               | 52 +++++-----
 sdgx/data_processors/transformers/base.py     | 20 ++--
 .../transformers/column_order.py              | 41 ++++----
 sdgx/data_processors/transformers/discrete.py | 98 ++++++++++---------
 sdgx/data_processors/transformers/nan.py      | 50 +++++-----
 sdgx/synthesizer.py                           |  4 +-
 tests/data_models/inspector/test_personal.py  | 11 ++-
 10 files changed, 180 insertions(+), 164 deletions(-)

diff --git a/sdgx/data_models/inspectors/personal.py b/sdgx/data_models/inspectors/personal.py
index 30b1ef97..3a8e84dd 100644
--- a/sdgx/data_models/inspectors/personal.py
+++ b/sdgx/data_models/inspectors/personal.py
@@ -73,13 +73,14 @@ def domain_verification(self, each_sample):
             return False
         return True
 
+
 class ChinaMainlandAddressInspector(RegexInspector):
 
-    # This regular expression does not take effect and is only for reference by developers. 
+    # This regular expression does not take effect and is only for reference by developers.
     # pattern = r"^[\u4e00-\u9fa5]{2,}(省|自治区|特别行政区|市)|[\u4e00-\u9fa5]{2,}(市|区|县|自治州|自治县|县级市|地区|盟|林区)?|[\u4e00-\u9fa5]{0,}(街道|镇|乡)?|[\u4e00-\u9fa5]{0,}(路|街|巷|弄)?|[\u4e00-\u9fa5]{0,}(号|弄)?$"
 
     pattern = r"^[\u4e00-\u9fa5]{2,}(省|自治区|特别行政区|市|县|村|弄|乡|路|街)"
-    
+
     pii = True
 
     data_type_name = "china_mainland_address"
@@ -93,10 +94,10 @@ def domain_verification(self, each_sample):
         if len(each_sample) > 30:
             return False
         # notice to distinguishing from the company name
-        if each_sample.endswith('公司'):
+        if each_sample.endswith("公司"):
             return False
         return True
-    
+
 
 @hookimpl
 def register(manager):
@@ -111,4 +112,3 @@ def register(manager):
     manager.register("ChinaMainlandUnifiedSocialCreditCode", ChinaMainlandUnifiedSocialCreditCode)
 
     manager.register("ChinaMainlandAddressInspector", ChinaMainlandAddressInspector)
-
diff --git a/sdgx/data_processors/formatters/base.py b/sdgx/data_processors/formatters/base.py
index 201df19d..83b810b9 100644
--- a/sdgx/data_processors/formatters/base.py
+++ b/sdgx/data_processors/formatters/base.py
@@ -1,6 +1,7 @@
 from __future__ import annotations
 
-import pandas as pd 
+import pandas as pd
+
 from sdgx.data_processors.base import DataProcessor
 
 
@@ -30,17 +31,16 @@ def reverse_convert(self, processed_data: pd.DataFrame) -> pd.DataFrame:
             pd.DataFrame: Raw data
         """
         return self.post_processing(processed_data)
-    
-     
+
     def post_processing(self, processed_data: pd.DataFrame) -> pd.DataFrame:
-        '''
+        """
         For formatter, please rewrite this method.
-        
+
         Args:
             processed_data (pd.DataFrame): Processed data
 
         Returns:
             pd.DataFrame: Raw data
-        '''
+        """
 
         return processed_data
diff --git a/sdgx/data_processors/formatters/int.py b/sdgx/data_processors/formatters/int.py
index 0ed4d35d..8e6e319f 100644
--- a/sdgx/data_processors/formatters/int.py
+++ b/sdgx/data_processors/formatters/int.py
@@ -1,49 +1,51 @@
 from __future__ import annotations
 
-import pandas as pd 
 from typing import Any
 
-from sdgx.data_models.metadata import Metadata  
-from sdgx.data_processors.formatters.base import Formatter
+import pandas as pd
+
+from sdgx.data_models.metadata import Metadata
 from sdgx.data_processors.extension import hookimpl
+from sdgx.data_processors.formatters.base import Formatter
 from sdgx.utils import logger
 
+
 class IntValueFormatter(Formatter):
-    '''
+    """
     Formatter class for handling Int values in pd.DataFrame.
-    '''
+    """
 
     int_columns = None
 
     def fit(self, metadata: Metadata | None = None, **kwargs: dict[str, Any]):
-        '''
-        Fit method for the formatter. 
-        
+        """
+        Fit method for the formatter.
+
         Formatter need to use metadata to record which columns belong to the int type, and convert them back to the int type during post-processing.
-        '''
-        
-        # get from metadata 
+        """
+
+        # get from metadata
         self.int_columns = metadata.get("int_columns")
-        
+
         logger.info("IntValueFormatter Fitted.")
 
-        return 
+        return
 
     def convert(self, raw_data: pd.DataFrame) -> pd.DataFrame:
-        '''
+        """
         Convert method to handle missing values in the input data.
-        '''
-        
+        """
+
         logger.info("Converting data using IntValueFormatter... Finished  (No Action).")
 
         return raw_data
-    
+
     def post_processing(self, processed_data: pd.DataFrame) -> pd.DataFrame:
-        '''
-        post_processing method for the formatter. 
-        
+        """
+        post_processing method for the formatter.
+
         Does not require any action.
-        '''
+        """
 
         for col in self.int_columns:
             processed_data[col] = processed_data[col].astype(int)
@@ -54,7 +56,7 @@ def post_processing(self, processed_data: pd.DataFrame) -> pd.DataFrame:
 
     pass
 
+
 @hookimpl
 def register(manager):
     manager.register("IntValueFormatter", IntValueFormatter)
-
diff --git a/sdgx/data_processors/manager.py b/sdgx/data_processors/manager.py
index 4c75cdfc..34105d00 100644
--- a/sdgx/data_processors/manager.py
+++ b/sdgx/data_processors/manager.py
@@ -8,11 +8,12 @@
 from sdgx.data_processors.extension import project_name as PROJECT_NAME
 from sdgx.manager import Manager
 
+
 class DataProcessorManager(Manager):
     """
     This is a plugin management class for data processing components.
-    
-    Properties: 
+
+    Properties:
         - register_type: Specifies the type of data processors to register.
         - project_name: Stores the project name from the extension module.
         - hookspecs_model: Stores the hook specifications model from the extension module.
@@ -27,63 +28,66 @@ class DataProcessorManager(Manager):
         - init_default_processors: Initializes default processors that are both registered and preset.
 
     """
+
     register_type = DataProcessor
-    
+
     project_name = PROJECT_NAME
-    
+
     hookspecs_model = extension
 
-    preset_defalut_processors = [ p.lower() for p in ["NonValueTransformer",'IntValueFormatter']]
-    
+    preset_defalut_processors = [p.lower() for p in ["NonValueTransformer", "IntValueFormatter"]]
+
     @property
     def registed_data_processors(self):
-        '''
+        """
         This property returns all registered data processors
-        '''
+        """
         return self.registed_cls
-    
+
     @property
     def registed_default_processor_list(self):
-        '''
+        """
         This property returns all registered default data processors
-        '''
+        """
         registed_processor_list = self.registed_data_processors.keys()
-        
-        # calculate intersection 
-        target_processors = list(set(registed_processor_list).intersection(self.preset_defalut_processors))
-        
+
+        # calculate intersection
+        target_processors = list(
+            set(registed_processor_list).intersection(self.preset_defalut_processors)
+        )
+
         return target_processors
 
     def load_all_local_model(self):
-        '''
+        """
         loads all local models
-        '''
+        """
         self._load_dir(data_processors.formatters)
         self._load_dir(data_processors.generators)
         self._load_dir(data_processors.samplers)
         self._load_dir(data_processors.transformers)
 
     def init_data_processor(self, processor_name, **kwargs: dict[str, Any]) -> DataProcessor:
-        '''
+        """
         Initializes a data processor with the given name and parameters
-        '''
+        """
         return self.init(processor_name, **kwargs)
 
     def init_all_processors(self, **kwargs: Any) -> list[DataProcessor]:
-        '''
+        """
         Initializes all registered data processors
-        '''
+        """
         return [
             self.init(processor_name, **kwargs)
             for processor_name in self.registed_data_processors.keys()
         ]
 
     def init_default_processors(self, **kwargs: Any) -> list[DataProcessor]:
-        '''
+        """
         Initializes all default data processors
-        '''
+        """
 
         return [
             self.init(processor_name, **kwargs)
             for processor_name in self.registed_default_processor_list
-        ]
\ No newline at end of file
+        ]
diff --git a/sdgx/data_processors/transformers/base.py b/sdgx/data_processors/transformers/base.py
index cd1860e2..5f40ff27 100644
--- a/sdgx/data_processors/transformers/base.py
+++ b/sdgx/data_processors/transformers/base.py
@@ -1,7 +1,8 @@
-import pandas as pd 
-from sdgx.data_processors.base import DataProcessor
-from sdgx.data_models.metadata import Metadata
+import pandas as pd
+
 from sdgx.data_loader import DataLoader
+from sdgx.data_models.metadata import Metadata
+from sdgx.data_processors.base import DataProcessor
 from sdgx.models.components.optimize.ndarray_loader import NDArrayLoader
 
 
@@ -16,19 +17,18 @@ class Transformer(DataProcessor):
     """
 
     def fit(self, metadata: Metadata | None = None, tabular_data: DataLoader | pd.DataFrame = None):
-        '''
-        Fit method for the transformer. 
-        '''
-        
+        """
+        Fit method for the transformer.
+        """
 
-        return  
+        return
 
     @staticmethod
-    def delete_column(tabular_data,  column_name):
+    def delete_column(tabular_data, column_name):
 
         pass
 
     @staticmethod
     def attach_columns(tabular_data, new_columns):
 
-        pass
\ No newline at end of file
+        pass
diff --git a/sdgx/data_processors/transformers/column_order.py b/sdgx/data_processors/transformers/column_order.py
index 82e6fbdb..253dbe47 100644
--- a/sdgx/data_processors/transformers/column_order.py
+++ b/sdgx/data_processors/transformers/column_order.py
@@ -1,50 +1,51 @@
 from __future__ import annotations
 
-from pandas import DataFrame  
-from sdgx.data_models.metadata import Metadata  
-from sdgx.data_processors.transformers.base import Transformer 
+from pandas import DataFrame
+
+from sdgx.data_models.metadata import Metadata
 from sdgx.data_processors.extension import hookimpl
+from sdgx.data_processors.transformers.base import Transformer
 from sdgx.utils import logger
 
+
 class ColumnOrderTransformer(Transformer):
-    '''
+    """
     Transformer class for handling missing values in data.
 
     This Transformer is mainly used as a reference for Transformer to facilitate developers to quickly understand the role of Transformer.
-    '''
+    """
 
     column_list: list = None
-    '''
+    """
     The list of tabular data's columns.
-    '''
+    """
 
     def fit(self, metadata: Metadata | None = None):
-        '''
-        Fit method for the transformer. 
-        
+        """
+        Fit method for the transformer.
+
         Remember the order of the columns.
-        '''
+        """
 
         self.column_list = list(metadata.column_list)
 
         logger.info("ColumnOrderTransformer Fitted.")
 
-        return 
+        return
 
     def convert(self, raw_data: DataFrame) -> DataFrame:
-        '''
+        """
         Convert method to handle missing values in the input data.
-        '''
+        """
         logger.info("Converting data using ColumnOrderTransformer...")
         logger.info("Converting data using ColumnOrderTransformer... Finished (No action).")
 
         return raw_data
-    
-    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
-        '''
-        Reverse_convert method for the transformer. 
-        '''
 
+    def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
+        """
+        Reverse_convert method for the transformer.
+        """
 
         logger.info("Data reverse-converted by ColumnOrderTransformer.")
 
@@ -52,7 +53,7 @@ def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
 
     pass
 
+
 @hookimpl
 def register(manager):
     manager.register("ColumnOrderTransformer", ColumnOrderTransformer)
-
diff --git a/sdgx/data_processors/transformers/discrete.py b/sdgx/data_processors/transformers/discrete.py
index b5e8e617..ad77621e 100644
--- a/sdgx/data_processors/transformers/discrete.py
+++ b/sdgx/data_processors/transformers/discrete.py
@@ -1,114 +1,116 @@
 from __future__ import annotations
 
-import pandas as pd 
+import pandas as pd
 from sklearn.preprocessing import OneHotEncoder
 
-from sdgx.data_models.metadata import Metadata  
-from sdgx.data_processors.transformers.base import Transformer 
-from sdgx.data_processors.extension import hookimpl
-from sdgx.utils import logger
 from sdgx.data_loader import DataLoader
+from sdgx.data_models.metadata import Metadata
+from sdgx.data_processors.extension import hookimpl
+from sdgx.data_processors.transformers.base import Transformer
 from sdgx.models.components.optimize.ndarray_loader import NDArrayLoader
+from sdgx.utils import logger
+
 
 class DiscreteTransformer(Transformer):
-    '''
+    """
     DiscreteTransformer is an important component of sdgx, used to handle discrete columns.
-    
+
     By default, DiscreteTransformer will perform one-hot encoding of discrete columns, and issue a warning message when dimensionality explosion occurs.
-    '''
+    """
 
     discrete_columns: list = None
-    '''
+    """
     Record which columns are of discrete type.
-    '''
+    """
 
     encoders: dict = {}
 
-    onehot_encoder_handle_unknown='ignore'
+    onehot_encoder_handle_unknown = "ignore"
 
     def fit(self, metadata: Metadata, tabular_data: DataLoader | pd.DataFrame):
-        '''
-        Fit method for the DiscreteTransformer. 
-        '''
+        """
+        Fit method for the DiscreteTransformer.
+        """
 
         logger.info("Fitting using DiscreteTransformer...")
 
-        self.discrete_columns = metadata.get('discrete_columns')
-        
-        # no discrete columns 
-        if len(self.discrete_columns) == 0 :
+        self.discrete_columns = metadata.get("discrete_columns")
+
+        # no discrete columns
+        if len(self.discrete_columns) == 0:
             logger.info("Fitting using DiscreteTransformer... Finished (No Columns).")
-            return 
+            return
 
         # then, there are >= 1 discrete colums
         for each_col in self.discrete_columns:
-            # fit each column 
+            # fit each column
             self._fit_column(each_col, tabular_data[[each_col]])
 
         logger.info("Fitting using DiscreteTransformer... Finished.")
-        
-        return 
-    
+
+        return
+
     def _fit_column(self, column_name: str, column_data: pd.DataFrame):
-        '''
+        """
         Fit every discrete columns in `_fit_column`.
 
         Args:
             - column_data (pd.DataFrame): A dataframe containing a column.
             - column_name: str: column name.
-        '''
+        """
 
-        self.encoders[column_name] = OneHotEncoder(handle_unknown= self.onehot_encoder_handle_unknown)
+        self.encoders[column_name] = OneHotEncoder(
+            handle_unknown=self.onehot_encoder_handle_unknown
+        )
         # fit the column data
         self.encoders[column_name].fit(column_data)
-        
+
         logger.info(f"Discrete column {column_name} fitted.")
-        
 
     def convert(self, raw_data: pd.DataFrame) -> pd.DataFrame:
-        '''
+        """
         Convert method to handle discrete values in the input data.
-        '''
+        """
 
         logger.info("Converting data using DiscreteTransformer...")
 
-        # TODO 
-        # transform every discrete column into 
+        # TODO
+        # transform every discrete column into
         if len(self.discrete_columns) == 0:
             logger.info("Converting data using DiscreteTransformer... Finished (No column).")
-            return 
-        
+            return
+
         for each_col in self.discrete_columns:
             new_onehot_column_set = self.encoders[each_col].transform(raw_data[[each_col]])
             # TODO 1- add new_onehot_column_set into the original dataframe
-            # TODO 2- delete the original column 
+            # TODO 2- delete the original column
             logger.info(f"Column {each_col} converted.")
-        
+
         logger.info("Converting data using DiscreteTransformer... Finished.")
-        
+
         # return the result
-        return 
-    
+        return
+
     def _transform_column(self, column_name: str, column_data: pd.DataFrame | pd.Series):
-        '''
+        """
         Transform every single discrete columns in `_transform_column`.
 
         Args:
             - column_data (pd.DataFrame): A dataframe containing a column.
             - column_name: str: column name.
 
-        '''
+        """
         pass
-    
+
     def reverse_convert(self, processed_data: pd.DataFrame) -> pd.DataFrame:
-        '''
-        Reverse_convert method for the transformer. 
-        
-        
-        '''
+        """
+        Reverse_convert method for the transformer.
+
+
+        """
 
         logger.info("Data reverse-converted by DiscreteTransformer.")
 
         return processed_data
 
-    pass
\ No newline at end of file
+    pass
diff --git a/sdgx/data_processors/transformers/nan.py b/sdgx/data_processors/transformers/nan.py
index ff07f8c1..053062ab 100644
--- a/sdgx/data_processors/transformers/nan.py
+++ b/sdgx/data_processors/transformers/nan.py
@@ -1,63 +1,65 @@
 from __future__ import annotations
 
-from pandas import DataFrame  
-from typing import Any 
+from typing import Any
 
-from sdgx.data_models.metadata import Metadata  
-from sdgx.data_processors.transformers.base import Transformer 
+from pandas import DataFrame
+
+from sdgx.data_models.metadata import Metadata
 from sdgx.data_processors.extension import hookimpl
+from sdgx.data_processors.transformers.base import Transformer
 from sdgx.utils import logger
 
+
 class NonValueTransformer(Transformer):
-    '''
+    """
     Transformer class for handling missing values in data.
 
     This Transformer is mainly used as a reference for Transformer to facilitate developers to quickly understand the role of Transformer.
-    '''
+    """
+
+    fill_na_value = 0
 
-    fill_na_value = 0 
-    
     drop_na = False
 
-    def fit(self, metadata: Metadata | None = None,  **kwargs: dict[str, Any]):
-        '''
-        Fit method for the transformer. 
-        
+    def fit(self, metadata: Metadata | None = None, **kwargs: dict[str, Any]):
+        """
+        Fit method for the transformer.
+
         Does not require any action.
-        '''
+        """
         logger.info("NonValueTransformer Fitted.")
 
-        return 
+        return
 
     def convert(self, raw_data: DataFrame) -> DataFrame:
-        '''
+        """
         Convert method to handle missing values in the input data.
-        '''
+        """
 
         logger.info("Converting data using NonValueTransformer...")
 
         if self.drop_na:
             res = raw_data.dropna()
         else:
-            res = raw_data.fillna(value= self.fill_na_value)  
-        
+            res = raw_data.fillna(value=self.fill_na_value)
+
         logger.info("Converting data using NonValueTransformer... Finished.")
 
         return res
-    
+
     def reverse_convert(self, processed_data: DataFrame) -> DataFrame:
-        '''
-        Reverse_convert method for the transformer. 
-        
+        """
+        Reverse_convert method for the transformer.
+
         Does not require any action.
-        '''
+        """
         logger.info("Data reverse-converted by NonValueTransformer (No Action).")
 
         return processed_data
 
     pass
 
+
 @hookimpl
 def register(manager):
     manager.register("NonValueTransformer", NonValueTransformer)
-
diff --git a/sdgx/synthesizer.py b/sdgx/synthesizer.py
index 0f38c72b..9b2daab8 100644
--- a/sdgx/synthesizer.py
+++ b/sdgx/synthesizer.py
@@ -291,7 +291,9 @@ def fit(
         start_time = time.time()
         for d in self.data_processors:
             d.fit(metadata)
-        logger.info(f"Fitted {len(self.data_processors)} data processors in  {time.time() - start_time}s.")
+        logger.info(
+            f"Fitted {len(self.data_processors)} data processors in  {time.time() - start_time}s."
+        )
 
         def chunk_generator() -> Generator[pd.DataFrame, None, None]:
             for chunk in self.dataloader.iter():
diff --git a/tests/data_models/inspector/test_personal.py b/tests/data_models/inspector/test_personal.py
index 1477c87d..9f417ac9 100644
--- a/tests/data_models/inspector/test_personal.py
+++ b/tests/data_models/inspector/test_personal.py
@@ -1,19 +1,20 @@
-import pytest
 import datetime
 import random
 import string
-import pandas as pd
 
+import pandas as pd
+import pytest
 from faker import Faker
+
 fake = Faker(locale="zh_CN")
 
 from sdgx.data_models.inspectors.personal import (
-    EmailInspector,
+    ChinaMainlandAddressInspector,
     ChinaMainlandIDInspector,
     ChinaMainlandMobilePhoneInspector,
     ChinaMainlandPostCode,
     ChinaMainlandUnifiedSocialCreditCode,
-    ChinaMainlandAddressInspector
+    EmailInspector,
 )
 
 
@@ -193,6 +194,7 @@ def test_chn_uscc_inspector_generated_data(chn_personal_test_df: pd.DataFrame):
     assert inspector_USCC.inspect_level == 30
     assert inspector_USCC.pii is True
 
+
 # CHN address
 def test_chn_address_inspector_demo_data(raw_data):
     inspector_CHN_Address = ChinaMainlandAddressInspector()
@@ -213,5 +215,6 @@ def test_chn_address_inspector_generated_data(chn_personal_test_df: pd.DataFrame
     assert inspector_CHN_Address.inspect_level == 30
     assert inspector_CHN_Address.pii is True
 
+
 if __name__ == "__main__":
     pytest.main(["-vv", "-s", __file__])