Update model.py

model.py

@@ -1,38 +1,39 @@
-import pandas as pd
-import joblib
-from keras.api.models import Sequential
-from keras.api.layers import Dense, Dropout, BatchNormalization
-from sklearn.preprocessing import StandardScaler
-
-# Veriyi yükle
-data = pd.read_csv("diabetes.csv")
-data = data[data.BMI >= 5]
-
-X = data.drop("Outcome", axis=1)
-y = data["Outcome"]
-
-# Ön işleme adımı
-sc = StandardScaler()
-X = sc.fit_transform(X)
-
-# Modeli oluşturma
-model = Sequential()
-model.add(Dense(120, activation="relu", input_shape=(X.shape[1],)))  # input_shape belirtin
-model.add(BatchNormalization())
-model.add(Dropout(0.3))  # Dropout oranı
-model.add(Dense(64, activation="relu"))
-model.add(BatchNormalization())
-model.add(Dropout(0.3))  # Dropout oranı
-model.add(Dense(1, activation="sigmoid"))  # Son katmanın çıkış boyutu 1 olmalı
-model.compile(loss="binary_crossentropy", optimizer="adam", metrics=["accuracy"])
-
-# Modeli eğitin
-model.fit(X, y,
-          epochs=10,
-          batch_size=32,
-          validation_split=0.2,
-          verbose=1)
-
-# Modeli kaydetme
-model.save("diabetes_model.keras")
-joblib.dump(sc, "diabetes_scaler.pkl")
+import pandas as pd
+import pickle
+from keras.api.models import Sequential
+from keras.api.layers import Dense, Dropout, BatchNormalization
+from sklearn.preprocessing import StandardScaler
+
+# Veriyi yükle
+data = pd.read_csv("diabetes.csv")
+data = data[data.BMI >= 5]
+
+X = data.drop("Outcome", axis=1)
+y = data["Outcome"]
+
+# Ön işleme adımı
+sc = StandardScaler()
+X = sc.fit_transform(X)
+
+# Modeli oluşturma
+model = Sequential()
+model.add(Dense(120, activation="relu", input_shape=(X.shape[1],)))  # input_shape belirtin
+model.add(BatchNormalization())
+model.add(Dropout(0.3))  # Dropout oranı
+model.add(Dense(64, activation="relu"))
+model.add(BatchNormalization())
+model.add(Dropout(0.3))  # Dropout oranı
+model.add(Dense(1, activation="sigmoid"))  # Son katmanın çıkış boyutu 1 olmalı
+model.compile(loss="binary_crossentropy", optimizer="adam", metrics=["accuracy"])
+
+# Modeli eğitin
+model.fit(X, y,
+          epochs=10,
+          batch_size=32,
+          validation_split=0.2,
+          verbose=1)
+
+# Modeli kaydetme
+model.save("model.keras")
+with open("dscaler.pkl", "wb") as scaler_file:
+    pickle.dump(scaler, scaler_file)