WIP: added base class for FMLearn algorithm

src/fmlearn.py

@@ -1,9 +1,63 @@
+import time
 import pandas as pd
 from sklearn.model_selection import train_test_split
 from sklearn.neighbors import KNeighborsClassifier
+from sklearn.metrics import accuracy_score
 
 from src.utils import utils
 
+class fmlearn:
+
+    def __init__(self):
+        self._X = None
+        self._y = None
+        self._model = None
+
+        # complete data frame used in fmlearn.
+        self._df = None
+        # stores the original shape of the dataframe before pre-processing
+        self._shape = None
+        # stores the time stamp of when the model was trained
+        self._model_create_time = None
+        self._accuracy = None
+
+    def load_data(self):
+        # loads data from the SQL database and pre-processes the data.
+        self._df = utils.get_df_from_db()
+        self._shape = self._df.shape
+
+        # replacing NA values in the dataframe with -1
+        self._df.fillna(-1, inplace=True)
+
+        self._X, self._y = utils.get_Xy(self._df)
+
+        # pre processing of data
+        self._X, _ = utils.ohe_feature(self._X, utils.TARGET_TYPE)
+
+        self._y, _ = utils.label_encode_feature(self._y, utils.DATASET_HASH)
+
+    def train(self):
+        if self._X is None:
+            print('data not loaded! \nloading data and then training model')
+            self.load_data()
+
+        X_train, X_test, y_train, y_test = train_test_split(self._X, self._y, test_size=0.2, random_state=123)
+
+        self._model = KNeighborsClassifier()
+        self._model.fit(X_train, y_train)
+
+        self._model_create_time = time.now()
+
+        y_pred = self._model.predict(X_test)
+        self._accuracy = accuracy_score(y_test, y_pred)
+
+    def predict(self, X_pred):
+        y_pred = self._model.predict(X_pred)
+        return y_pred
+
+
+
+
 def kmc():
     df = utils.get_df_from_db()
     df.fillna(-1, inplace=True)
@@ -24,6 +78,7 @@ def kmc():
 
     y_pred = model.predict(X_test)
 
+    print(model.kneighbors(X_test)[1])
     print(y_test.to_string(header=False))
     y_pred = pd.DataFrame(y_pred)
     return y_pred.to_string(header=False)