add mlflow

MLFLOW/MLproject

@@ -0,0 +1,10 @@
+name: s464980
+
+docker_env:
+  image:  s464980-mlflow
+
+entry_points:
+  main:
+    parameters:
+      epochs: float
+    command: "python train.py --epochs {epochs}"

MLFLOW/train.py

@@ -0,0 +1,58 @@
+import pandas as pd
+from tensorflow import keras
+from tensorflow.keras import layers
+import argparse
+import mlflow
+
+
+class RegressionModel:
+    def __init__(self, optimizer="adam", loss="mean_squared_error"):
+        self.model = keras.Sequential([
+            layers.Input(shape=(5,)),  # Input layer
+            layers.Dense(32, activation='relu'),  # Hidden layer with 32 neurons and ReLU activation
+            layers.Dense(1)  # Output layer with a single neuron (for regression)
+        ])
+        self.optimizer = optimizer
+        self.loss = loss
+        self.X_train = None
+        self.X_test = None
+        self.y_train = None
+        self.y_test = None
+
+    def load_data(self, train_path, test_path):
+        data_train = pd.read_csv(train_path)
+        data_test = pd.read_csv(test_path)
+        self.X_train = data_train.drop("Performance Index", axis=1)
+        self.y_train = data_train["Performance Index"]
+        self.X_test = data_test.drop("Performance Index", axis=1)
+        self.y_test = data_test["Performance Index"]
+
+    def train(self, epochs=30):
+        self.model.compile(optimizer=self.optimizer, loss=self.loss)
+        self.model.fit(self.X_train, self.y_train, epochs=epochs, batch_size=32, validation_data=(self.X_test, self.y_test))
+
+    def predict(self, data):
+        prediction = self.model.predict(data)
+        return prediction
+
+    def evaluate(self):
+        test_loss = self.model.evaluate(self.X_test, self.y_test)
+        print(f"Test Loss: {test_loss:.4f}")
+        return test_loss
+
+    def save_model(self):
+        self.model.save("model.keras")
+
+
+parser = argparse.ArgumentParser()
+parser.add_argument('--epochs')
+
+args = parser.parse_args()
+mlflow.set_tracking_uri("http://localhost:5000")
+model = RegressionModel()
+with mlflow.start_run() as run:
+    model.train(epochs=int(args.epochs))
+    rmse = model.evaluate()
+    mlflow.log_param("epoch", int(args.epochs))
+    mlflow.log_metric("rmse", rmse)
+model.save_model()

MLproject

@@ -1,9 +0,0 @@
-name: s464980
-
-conda_env: conda.yaml
-
-entry_points:
-  optimal_parameters:
-    parameters:
-      epochs: { type: int, default: 20 }
-    command: 'python train.py {epochs}'

conda.yaml

@@ -1,13 +0,0 @@
-name: s464980
-channels:
-  - defaults
-dependencies:
-  - python=3.11
-  - pip
-  - pip:
-      - mlflow
-      - tensorflow
-      - pandas
-      - scikit-learn
-      - numpy
-      - matplotlib

train.py

@@ -2,8 +2,6 @@ import pandas as pd
 from tensorflow import keras
 from tensorflow.keras import layers
 import argparse
-import mlflow
-
 
 class RegressionModel:
     def __init__(self, optimizer="adam", loss="mean_squared_error"):
@@ -48,11 +46,7 @@ parser = argparse.ArgumentParser()
 parser.add_argument('--epochs')
 
 args = parser.parse_args()
-mlflow.set_tracking_uri("http://localhost:5000")
 model = RegressionModel()
-with mlflow.start_run() as run:
+model.load_data("df_train.csv", "df_test.csv")
-    model.train(epochs=int(args.epochs))
+model.train(epochs=int(args.epochs))
-    rmse = model.evaluate()
-    mlflow.log_param("epoch", int(args.epochs))
-    mlflow.log_metric("rmse", rmse)
 model.save_model()