Training jenkinsfile

Jenkinsfile

@@ -34,6 +34,7 @@ pipeline {
 				sh "./download_data.sh"
 				sh "python3 clean_and_split_data.py"  
 				archiveArtifacts artifacts: "crime_test.csv, crime_dev.csv, crime_train.csv", onlyIfSuccessful: true               
+			    build job: "s444380-training/master"
 			}
         }
     }

Jenkinsfile.train

@@ -0,0 +1,44 @@
+pipeline {
+    agent {
+		docker { image 'kamilguttmann/ium:latest' }
+	}
+    parameters {
+        string (
+                defaultValue: '100',
+                description: 'Epochs',
+                name: 'EPOCHS'
+            )
+        buildSelector(
+            defaultSelector: lastSuccessful(),
+            description: 'Which build to use for copying artifacts',
+            name: 'BUILD_SELECTOR'
+        )
+    }
+    stages {
+        stage('Copy artifacts') {
+            steps {
+                copyArtifacts fingerprintArtifacts: true, projectName: 's444380-create-dataset', selector: buildParameter('BUILD_SELECTOR')
+            }
+        }
+        stage("Checkout git") {
+            steps {
+	        checkout([$class: 'GitSCM', branches: [[name: '*/master']], extensions: [], userRemoteConfigs: [[credentialsId: 's444380', url: 'https://git.wmi.amu.edu.pl/s444380/ium_444380.git']]])
+            }
+        }
+        stage("Train model") {
+            steps {
+                sh "./train_model.py $EPOCHS"
+                archiveArtifacts artifacts: "./model/*", onlyIfSuccessful: true
+            }
+        }
+    }
+    post {
+        success {
+            emailtext body: "SUCCESS", subject: "s444380-training", to: "e19191c5.uam.onmicrosoft.com@emea.teams.ms"
+        }
+
+        failure {
+            emailtext body: "FAILURE", subject: "s444380-training", to: "e19191c5.uam.onmicrosoft.com@emea.teams.ms"
+        }
+    }
+}

train_model.py

@@ -1,3 +1,5 @@
+import sys
+
 import pandas as pd
 from sklearn.preprocessing import LabelEncoder
 from keras.models import Sequential
@@ -63,8 +65,13 @@ model.add(Dense(128, activation='relu'))
 model.add(Dense(num_categories, activation='softmax'))
 model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy', 'sparse_categorical_accuracy'])
 
+if len(sys.argv) > 1:
+    epochs = sys.argv[1]
+else:
+    epochs = 10
+
 # Train model
-model.fit(x_train, y_train, epochs=100, validation_data=(x_val, y_val))
+model.fit(x_train, y_train, epochs=epochs, validation_data=(x_val, y_val))
 
 # Make predictions
 y_pred = model.predict(x_test)