lab5 fix

Jenkinsfile

@@ -2,6 +2,13 @@ pipeline {
     agent {
         dockerfile true
         }
+	parameters {
+        string(
+			defaultValue: '100',
+			description: 'Example training parameter',
+			name: 'EPOCHS_NUM'
+		)
+    }
     stages {
         stage('Stage 1') {
             steps {
@@ -10,7 +17,7 @@ pipeline {
                     sh 'python3 process_dataset.py'
 					echo 'Dataset processed'
 					echo 'Conducting simple regression model test'
-					sh 'python3 simple_regression.py'
+					sh 'python3 simple_regression.py $EPOCHS_NUM'
 					echo 'Model predictions saved'
                     sh 'head lego_linreg_results.csv'
 				}

simple_regression.py

@@ -44,18 +44,11 @@ history = model.fit(
     validation_split=0.2
 )
 
-# Prosta ewaluacja
+# Wykonanie predykcji na danych ze zbioru testującego
-test_results = {'model': model.evaluate(
+y_pred = model.predict(test_piece_counts)
-    test_piece_counts,
-    test_prices, verbose=0)
-}
-
-# Wykonanie wielu predykcji
-x = tf.linspace(100, 7000, 6901)
-y = model.predict(x)
 
 # Zapis predykcji do pliku
-results = pd.DataFrame({"input_piece_count": x.numpy().tolist(), "predicted_price": [a[0] for a in y.tolist()]})
+results = pd.DataFrame({"test_set_piece_count": test_piece_counts.numpy().tolist(), "predicted_price": [a[0] for a in y_pred.tolist()]})
 results.to_csv(r'lego_reg_results.csv', index=False, header=True)
 
 # Zapis modelu do pliku
@@ -63,6 +56,12 @@ model.save('lego_reg_model')
 
 # Opcjonalne statystyki, wykresy
 '''
+# Prosta ewaluacja
+test_results = {'model': model.evaluate(
+    test_piece_counts,
+    test_prices, verbose=0)
+}
+
 print(test_results)
 
 hist = pd.DataFrame(history.history)
@@ -70,7 +69,7 @@ hist['epoch'] = history.epoch
 print(hist.tail())
 
 plt.scatter(train_piece_counts, train_prices, label='Data')
-plt.plot(x, y, color='k', label='Predictions')
+plt.plot(x, y_pred, color='k', label='Predictions')
 plt.xlabel('pieces')
 plt.ylabel('price')
 plt.legend()