mlflow

Jenkins/Jenkinsfile.create_dataset

@@ -1,6 +1,6 @@
 node {
     checkout scm
-    def dockerImageIUM = docker.build("s444452/ium:1.3")
+    def dockerImageIUM = docker.build("s444452/ium:1.4")
     dockerImageIUM.inside {
         stage('Preparation') {
             properties([

Jenkins/Jenkinsfile.evaluation

@@ -1,7 +1,7 @@
 node {
     checkout scm
     try {
-        docker.image('s444452/ium:1.3').inside {
+        docker.image('s444452/ium:1.4').inside {
         stage('Preparation') {
             properties([
                 parameters([

Jenkins/Jenkinsfile.stats

@@ -1,5 +1,5 @@
 node {
-    docker.image('s444452/ium:1.3').inside {
+    docker.image('s444452/ium:1.4').inside {
         stage('Preparation') {
         properties([parameters([
             buildSelector(

Jenkins/Jenkinsfile.training

@@ -1,7 +1,7 @@
 node {
     checkout scm
     try {
-        docker.image('s444452/ium:1.3').inside {
+        docker.image('s444452/ium:1.4').inside {
         stage('Preparation') {
             properties([
                 pipelineTriggers([upstream(threshold: hudson.model.Result.SUCCESS, upstreamProjects: "s444452-create-dataset")]),

MLproject

@@ -0,0 +1,23 @@
+name: s444452_fake_job_classification
+
+docker_env:
+    image: s444452/ium:1.4
+
+entry_points:
+    train:
+        parameters:
+            data_path: path
+            epochs: {type: float, default: 1}
+            num_words: {type: float, default: 20000}
+            batch_size: {type: float, default: 150}
+            pad_length: {type: float, default: 300}
+        command: "python3 Scripts/train_neural_network.py {data_path} {epochs} {num_words} {batch_size} {pad_length}"
+    evaluate:
+        parameters:
+            build_nr: {type: float, default: 0}
+            data_path: path
+            epochs: {type: float, default: 1}
+            num_words: {type: float, default: 20000}
+            batch_size: {type: float, default: 150}
+            pad_length: {type: float, default: 300}
+        command: "python3 Scripts/evaluate_neural_network.py {build_nr} {data_path} {epochs} {num_words} {batch_size} {pad_length}"

Scripts/evaluate_neural_network.py

@@ -11,6 +11,14 @@ import matplotlib.pyplot as plt
 from sacred.observers import MongoObserver
 from sacred.observers import FileStorageObserver
 from sacred import Experiment
+import mlflow
+import logging
+
+logging.basicConfig(level=logging.WARN)
+logger = logging.getLogger(__name__)
+
+mlflow.set_tracking_uri("http://172.17.0.1:5000")
+mlflow.set_experiment("s444452")
 
 ex = Experiment(name='s444452_fake_job_classification_evaluation', save_git_info=False)
 ex.observers.append(MongoObserver(url='mongodb://admin:IUM_2021@172.17.0.1:27017',
@@ -51,8 +59,16 @@ def evaluate_and_save(model, x, y, abs_path, build_number, _run):
     y_predicted = (model.predict(x) >= 0.5).astype(int)
     evaluation_file_path = os.path.join(abs_path, 'neural_network_evaluation.csv')
     with open(evaluation_file_path, 'a+') as f:
-        result = f'{build_number},{accuracy},{loss},{precision_score(y, y_predicted)},{recall_score(y, y_predicted)},{f1_score(y, y_predicted)}'
+        precision = precision_score(y, y_predicted)
+        recall = recall_score(y, y_predicted)
+        f1 = f1_score(y, y_predicted)
+        result = f'{build_number},{accuracy},{loss},{precision},{recall},{f1}'
         f.write(result + '\n')
+        mlflow.log_metric("accuracy", accuracy)
+        mlflow.log_metric("loss", loss)
+        mlflow.log_metric("precision", precision)
+        mlflow.log_metric("recall", recall)
+        mlflow.log_metric("f1_score", f1)
     # ex.log_scalar("loss", loss)
     _run.log_scalar("training.loss", loss)
     # ex.log_scalar("accuracy", accuracy)
@@ -102,6 +118,16 @@ def load_data(data_path, filename) -> pd.DataFrame:
 
 @ex.main
 def main(build_number, data_path, num_words, epochs, batch_size, pad_length, _run):
+    with mlflow.start_run() as mlflow_run:
+        print("MLflow run experiment_id: {0}".format(mlflow_run.info.experiment_id))
+        print("MLflow run artifact_uri: {0}".format(mlflow_run.info.artifact_uri))
+        mlflow.log_param("build_number", build_number)
+        mlflow.log_param("data_path", data_path)
+        mlflow.log_param("num_words", num_words)
+        mlflow.log_param("epochs", epochs)
+        mlflow.log_param("batch_size", batch_size)
+        mlflow.log_param("pad_length", pad_length)
+
         abs_data_path = os.path.abspath(data_path)
         train_data = load_data(abs_data_path, 'train_data.csv')
         test_data = load_data(abs_data_path, 'test_data.csv')

Scripts/train_neural_network.py

@@ -1,6 +1,9 @@
 #!/usr/bin/python
 import os
 import sys
+import warnings
+
+import numpy as np
 import pandas as pd
 from keras.models import Sequential
 from keras import layers
@@ -9,6 +12,15 @@ from keras.preprocessing.sequence import pad_sequences
 from sacred.observers import MongoObserver
 from sacred.observers import FileStorageObserver
 from sacred import Experiment
+from mlflow.models.signature import infer_signature
+import mlflow
+import logging
+
+logging.basicConfig(level=logging.WARN)
+logger = logging.getLogger(__name__)
+
+mlflow.set_tracking_uri("http://172.17.0.1:5000")
+mlflow.set_experiment("s444452")
 
 ex = Experiment(name='s444452_fake_job_classification_training', save_git_info=False)
 ex.observers.append(MongoObserver(url='mongodb://admin:IUM_2021@172.17.0.1:27017',
@@ -80,6 +92,15 @@ def load_data(data_path, filename) -> pd.DataFrame:
 
 @ex.main
 def main(data_path, num_words, epochs, batch_size, pad_length, _run):
+    with mlflow.start_run() as mlflow_run:
+        print("MLflow run experiment_id: {0}".format(mlflow_run.info.experiment_id))
+        print("MLflow run artifact_uri: {0}".format(mlflow_run.info.artifact_uri))
+        mlflow.log_param("data_path", data_path)
+        mlflow.log_param("num_words", num_words)
+        mlflow.log_param("epochs", epochs)
+        mlflow.log_param("batch_size", batch_size)
+        mlflow.log_param("pad_length", pad_length)
+
         abs_data_path = os.path.abspath(data_path)
         train_data = load_data(abs_data_path, 'train_data.csv')
         test_data = load_data(abs_data_path, 'test_data.csv')
@@ -90,5 +111,10 @@ def main(data_path, num_words, epochs, batch_size, pad_length, _run):
         train_model(model, x_train, y_train)
         save_model(model)
 
+        signature = infer_signature(x_train, y_train)
+        input_example = np.array(x_test[:20])
+        mlflow.keras.log_model(model, "model", signature=signature, input_example=input_example)
 
+
+warnings.filterwarnings("ignore")
 ex.run()

requirements.txt

@@ -1,12 +1,18 @@
 absl-py==1.0.0
+alembic==1.7.7
 astunparse==1.6.3
 cachetools==5.0.0
 certifi==2021.10.8
 charset-normalizer==2.0.12
 click==8.1.2
+cloudpickle==2.0.0
 colorama==0.4.4
 cycler==0.11.0
+databricks-cli==0.16.6
+docker==5.0.3
 docopt==0.6.2
+entrypoints==0.4
+Flask==2.1.2
 flatbuffers==2.0
 fonttools==4.33.3
 gast==0.5.3
@@ -15,10 +21,14 @@ GitPython==3.1.27
 google-auth==2.6.6
 google-auth-oauthlib==0.4.6
 google-pasta==0.2.0
+greenlet==1.1.2
 grpcio==1.44.0
+gunicorn==20.1.0
 h5py==3.6.0
 idna==3.3
 importlib-metadata==4.11.3
+itsdangerous==2.1.2
+Jinja2==3.1.2
 joblib==1.1.0
 jsonpickle==1.5.2
 kaggle==1.5.12
@@ -26,8 +36,11 @@ keras==2.8.0
 Keras-Preprocessing==1.1.2
 kiwisolver==1.4.2
 libclang==14.0.1
+Mako==1.2.0
 Markdown==3.3.6
+MarkupSafe==2.1.1
 matplotlib==3.5.2
+mlflow==1.25.1
 munch==2.5.0
 nltk==3.7
 numpy==1.22.3
@@ -36,15 +49,20 @@ opt-einsum==3.3.0
 packaging==21.3
 pandas==1.4.2
 Pillow==9.1.0
+prometheus-client==0.14.1
+prometheus-flask-exporter==0.20.1
 protobuf==3.20.1
 py-cpuinfo==8.0.0
 pyasn1==0.4.8
 pyasn1-modules==0.2.8
+PyJWT==2.3.0
 pymongo==4.1.1
 pyparsing==3.0.8
 python-dateutil==2.8.2
 python-slugify==6.1.1
 pytz==2022.1
+PyYAML==6.0
+querystring-parser==1.2.4
 regex==2022.3.15
 requests==2.27.1
 requests-oauthlib==1.3.1
@@ -55,6 +73,9 @@ scipy==1.8.0
 six==1.16.0
 sklearn==0.0
 smmap==5.0.0
+SQLAlchemy==1.4.36
+sqlparse==0.4.2
+tabulate==0.8.9
 tensorboard==2.8.0
 tensorboard-data-server==0.6.1
 tensorboard-plugin-wit==1.8.1
@@ -67,6 +88,7 @@ threadpoolctl==3.1.0
 tqdm==4.64.0
 typing_extensions==4.2.0
 urllib3==1.26.9
+websocket-client==1.3.2
 Werkzeug==2.1.1
 wrapt==1.14.0
 zipp==3.8.0