From 9d66acbdad756b441f7d465c3478a1689c9614a0 Mon Sep 17 00:00:00 2001
From: zgolebiewska <zgolebiewska5@gmail.com>
Date: Sun, 26 May 2024 14:16:50 +0200
Subject: [PATCH] IUM_08

---
 PMProject.yaml | 14 ++++++++++++++
 conda.yaml     | 10 ++++++++++
 model.py       | 21 ++++++++++++++++-----
 3 files changed, 40 insertions(+), 5 deletions(-)
 create mode 100644 PMProject.yaml
 create mode 100644 conda.yaml

diff --git a/PMProject.yaml b/PMProject.yaml
new file mode 100644
index 0000000..0ed480e
--- /dev/null
+++ b/PMProject.yaml
@@ -0,0 +1,14 @@
+name: OrangeQualityModel
+
+conda_env: conda.yaml
+
+entry_points:
+  train:
+    parameters:
+      epochs: {type: int, default: 100}
+    command: "python model.py"
+
+  test:
+    parameters:
+      model_path: {type: str, default: "orange_quality_model_tf.h5"}
+    command: "python test_model.py --model_path {model_path}"
diff --git a/conda.yaml b/conda.yaml
new file mode 100644
index 0000000..f759448
--- /dev/null
+++ b/conda.yaml
@@ -0,0 +1,10 @@
+name: orange_quality_model_env
+channels:
+  - defaults
+dependencies:
+  - python=3.8
+  - pip
+  - pandas
+  - scikit-learn
+  - tensorflow
+  - mlflow
\ No newline at end of file
diff --git a/model.py b/model.py
index efa1509..f1633c8 100644
--- a/model.py
+++ b/model.py
@@ -3,6 +3,9 @@ import pandas as pd
 from sklearn.preprocessing import LabelEncoder, StandardScaler
 from sklearn.model_selection import train_test_split
 import json
+import mlflow
+
+mlflow.set_tracking_uri("http://localhost:5000")  # Ustawienie adresu MLflow Tracking Server
 
 df = pd.read_csv('OrangeQualityData.csv')
 
@@ -30,11 +33,19 @@ model = tf.keras.Sequential([
 
 model.compile(optimizer='sgd', loss='mse')
 
-history = model.fit(X_train_scaled, y_train, epochs=100, verbose=0, validation_data=(X_test_scaled, y_test))
+with mlflow.start_run():
+    mlflow.log_param("optimizer", 'sgd')
+    mlflow.log_param("loss_function", 'mse')
+    mlflow.log_param("epochs", 100)
 
-model.save('orange_quality_model_tf.h5')
+    history = model.fit(X_train_scaled, y_train, epochs=100, verbose=0, validation_data=(X_test_scaled, y_test))
 
-predictions = model.predict(X_test_scaled)
+    for key, value in history.history.items():
+        mlflow.log_metric(key, value[-1])  # Logujemy ostatnią wartość metryki
 
-with open('predictions_tf.json', 'w') as f:
-    json.dump(predictions.tolist(), f, indent=4)
+    model.save('orange_quality_model_tf.h5')
+
+    predictions = model.predict(X_test_scaled)
+
+    with open('predictions_tf.json', 'w') as f:
+        json.dump(predictions.tolist(), f, indent=4)