fix

Dockerfile

@@ -2,7 +2,7 @@ FROM ubuntu:latest
 
 RUN apt-get update && apt-get install -y python3-pip unzip coreutils
 
-RUN pip install --user kaggle pandas scikit-learn tensorflow
+RUN pip install --no-cache-dir wheel kaggle pandas scikit-learn tensorflow
 
 WORKDIR /app
 

Jenkinsfile

@@ -1,36 +1,61 @@
 pipeline {
-    agent any
+   agent {
+    dockerfile true
+    }
+
+   triggers {
+     upstream(upstreamProjects: 's464937-training/training', threshold: hudson.model.Result.SUCCESS)
+   }
 
    parameters {
-        string(name: 'CUTOFF', defaultValue: '100', description: 'Ilość wierszy do odcięcia')
-	    string(name: 'KAGGLE_USERNAME', defaultValue: '', description: 'Kaggle username')
-        password(name: 'KAGGLE_KEY', defaultValue: '', description: 'Kaggle API key')
+     buildSelector(defaultSelector: lastSuccessful(), description: 'Which build to use for copying artifacts', name: 'BUILD_SELECTOR')
+     gitParameter branchFilter: 'origin/(.*)', defaultValue: 'training', name: 'BRANCH', type: 'PT_BRANCH'
    }
 
     stages {
-        stage('Clone repo') {
+        stage('Clone Repository') {
             steps {
-                git branch: "main", url: "https://git.wmi.amu.edu.pl/s464937/ium_464937"
+                git branch: 'evaluation', url: "https://git.wmi.amu.edu.pl/s464937/ium_464937"
+            }
+        }
+        stage('Copy Dataset Artifacts') {
+            steps {
+                 copyArtifacts filter: 'data/dev.csv,data/test.csv,data/train.csv', projectName: 'z-s464937-create-dataset', selector: buildParameter('BUILD_SELECTOR')
+            }
+        }
+        stage('Copy Training Artifacts') {
+          steps {
+            copyArtifacts filter: 'powerlifting_model.h5', projectName: 's464937-training/' + params.BRANCH, selector: buildParameter('BUILD_SELECTOR')
+          }
+        }
+        stage('Copy Evaluation Artifacts') {
+          steps {
+            copyArtifacts filter: 'metrics.txt', projectName: 's464937-evaluation/evaluation', selector: buildParameter('BUILD_SELECTOR'), optional: true
+          }
+        }
+        stage("Run predictions") {
+            steps {
+                sh "chmod +x ./predict.py"
+                sh "python3 ./predict.py"
+                archiveArtifacts artifacts: 'powerlifting_test_predictions.csv', onlyIfSuccessful: true
+            }
+        }
+        stage('Run metrics') {
+          steps {
+            sh 'chmod +x ./metrics.py'
+            sh "python3 ./metrics.py ${currentBuild.number}"
           }
         }
 
-        stage('Download and preprocess') {
-            environment {
-                    KAGGLE_USERNAME = "szymonbartanowicz"
-                    KAGGLE_KEY = "4692239eb65f20ec79f9a59ef30e67eb"
-                }
+        stage('Run plot') {
           steps {
-                withEnv([
-                    "KAGGLE_USERNAME=${env.KAGGLE_USERNAME}",
-                    "KAGGLE_KEY=${env.KAGGLE_KEY}"
-                ]) {
-                    sh "bash ./script1.sh ${params.CUTOFF}"
+            sh 'chmod +x ./plot.py'
+            sh 'python3 ./plot.py'
           }
         }
-        }
-        stage('Archive') {
+         stage('Archive Artifacts') {
           steps {
-                archiveArtifacts artifacts: 'data/*', onlyIfSuccessful: true
+            archiveArtifacts artifacts: '*', onlyIfSuccessful: true
           }
         }
     }

metrics.py

@@ -0,0 +1,14 @@
+import pandas as pd
+from sklearn.metrics import accuracy_score, precision_recall_fscore_support, mean_squared_error
+from math import sqrt
+import sys
+
+data = pd.read_csv('powerlifting_test_predictions.csv')
+y_pred = data['predicted_TotalKg']
+y_test = data['actual_TotalKg']
+
+build_number = sys.argv[1]
+rmse = sqrt(mean_squared_error(y_test, y_pred))
+
+with open(r"metrics.txt", "a") as f:
+    f.write(f"{build_number},{rmse}\n")

plot.py

@@ -0,0 +1,20 @@
+import matplotlib.pyplot as plt
+
+def main():
+    accuracy = []
+    build_numbers = []
+
+    with open("metrics.txt") as f:
+        for line in f:
+            accuracy.append(float(line.split(",")[0]))
+            build_numbers.append(int(line.split(",")[1]))
+
+    plt.plot(build_numbers, accuracy)
+    plt.xlabel("Build Number")
+    plt.ylabel("RMSE")
+    plt.title("RMSE of the model over time")
+    plt.xticks(range(min(build_numbers), max(build_numbers) + 1))
+    plt.savefig("plot.png")
+
+if __name__ == "__main__":
+    main()

predict.py

@@ -4,11 +4,18 @@ from sklearn.preprocessing import StandardScaler, OneHotEncoder
 from sklearn.compose import ColumnTransformer
 from sklearn.pipeline import Pipeline
 from sklearn.model_selection import train_test_split
+from keras.metrics import MeanSquaredError
 
 loaded_model = tf.keras.models.load_model('powerlifting_model.h5')
 
 data = pd.read_csv('openpowerlifting.csv')
+
 data = data[['Sex', 'Age', 'BodyweightKg', 'TotalKg']].dropna()
+
+data['Age'] = pd.to_numeric(data['Age'], errors='coerce')
+data['BodyweightKg'] = pd.to_numeric(data['BodyweightKg'], errors='coerce')
+data['TotalKg'] = pd.to_numeric(data['TotalKg'], errors='coerce')
+
 features = data[['Sex', 'Age', 'BodyweightKg']]
 target = data['TotalKg']
 
@@ -20,8 +27,9 @@ preprocessor = ColumnTransformer(
         ('cat', OneHotEncoder(), ['Sex'])
     ]
 )
-X_test_transformed = preprocessor.fit_transform(X_test)
 
+X_test_transformed = preprocessor.fit_transform(X_test)
 predictions = loaded_model.predict(X_test_transformed)
 predictions_df = pd.DataFrame(predictions, columns=['predicted_TotalKg'])
+predictions_df['actual_TotalKg'] = y_test.reset_index(drop=True)
 predictions_df.to_csv('powerlifting_test_predictions.csv', index=False)