Merge pull request #29 from snega16/snega16

Thanks, Good Work. Merged!

tab_automl/automl/datasets.py

@@ -106,6 +106,36 @@ def prepare_x_and_y(self,
         print(f"X feature set and target feature has been split...")
 
         return x, y
+
+
+class ClusteringDataset:
+    """
+
+    Dataset Class for Clustering.
+    Currently supports single feature target only.
+
+    """
+
+    def __init__(self, path):
+        # reading the dataset from source
+        if path.endswith(".txt"):
+            self.data = pd.read_table(path, delimiter='\s')
+        elif path.endswith(".json"):
+            self.data = pd.read_json(path)
+        elif path.endswith(".xlsx"):
+            self.data = pd.read_excel(path)
+        elif path.endswith(".sqlite"):
+            table_name = input("table name :")
+            db = sqlite3.connect(path)
+            self.data = pd.read_sql_query(f'Select * from {table_name}', db)
+        elif path.endswith(".csv"):
+            self.data = pd.read_csv(path)
+        else:
+            print("File type not supported")
+        # storing the columns overview
+        self.columns = pd.Series([str(self.data[feature].dtype) for feature in self.data.columns])
+        self.labels = None
+        print(f"Populated the dataframe with data records...")
 
 
 class Iris:
@@ -190,3 +220,35 @@ def prepare_x_and_y(self,
         print(f"X feature set and target feature has been split...")
 
         return x, y
+
+
+class Credit_Card_Customer_Data:
+    """
+
+    Credit_Card_Customer_Data Dataset
+   ============
+
+    This class will be an example for
+    other clustering datasets or for
+    custom datasets prepared on comma
+    separated value format to be precise.
+
+    """
+
+    def __init__(self):
+        # reading the dataset from source
+        self.data = pd.read_csv("tab_automl/datasets/Credit_Card_Customer_Data.csv")
+        # storing the columns overview
+        self.columns = pd.Series([str(self.data[feature].dtype) for feature in self.data.columns])
+        self.labels = None
+        print(f"Populated the dataframe with data records...")
+
+        """
+        Defining x
+        """
+
+        x = self.data
+        print(f"X feature set has been set...")
+
+        return x
+