diff --git a/.github/workflows/pmml.yml b/.github/workflows/pmml.yml
index 68eb0ef..27313f7 100644
--- a/.github/workflows/pmml.yml
+++ b/.github/workflows/pmml.yml
@@ -41,3 +41,37 @@ jobs:
     - name: Run All Steps (1-3)
       run: |
         python run_all_steps.py
+
+    - name: Run tests
+      run: |
+        python utils/metrics_test.py
+
+     shell: bash
+      run: |
+        accuracy=$(grep -Po 'accuracy_score: \K[0-9.]+' Metrics_output.txt)
+        mae=$(grep -Po 'MAE: \K[0-9.]+' Metrics_output.txt)
+        echo "Accuracy: $accuracy"
+        echo "MAE: $mae"
+        echo "::set-output name=accuracy::$accuracy"
+        echo "::set-output name=mae::$mae"
+
+    - name: Post results to PR
+      if: github.event_name == 'pull_request'
+      uses: actions/github-script@v6
+      with:
+        script: |
+          const accuracy = "${{ steps.compute_metrics.outputs.accuracy }}";
+          const mae = "${{ steps.compute_metrics.outputs.mae }}";
+          const body = `
+          ### Metrics from latest run:
+          - **Accuracy**: ${accuracy}
+          - **MAE**: ${mae}
+          `;
+
+          github.rest.issues.createComment({
+            issue_number: context.issue.number,
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            body: body
+          });
+          
diff --git a/pmml/Metrics_output.txt b/pmml/Metrics_output.txt
new file mode 100644
index 0000000..219a18f
--- /dev/null
+++ b/pmml/Metrics_output.txt
@@ -0,0 +1,2 @@
+accuracy_score: 0.75
+MAE: 0.3333333333333333
diff --git a/pmml/utils/Metrics_output.txt b/pmml/utils/Metrics_output.txt
new file mode 100644
index 0000000..219a18f
--- /dev/null
+++ b/pmml/utils/Metrics_output.txt
@@ -0,0 +1,2 @@
+accuracy_score: 0.75
+MAE: 0.3333333333333333
diff --git a/pmml/utils/metrics_test.py b/pmml/utils/metrics_test.py
new file mode 100644
index 0000000..752cc74
--- /dev/null
+++ b/pmml/utils/metrics_test.py
@@ -0,0 +1,22 @@
+from sklearn.metrics import accuracy_score, mean_absolute_error
+import pandas
+from os.path import basename
+
+# Load the preprocessed test data CSV into a DataFrame
+storybooks_csv_path = '../step1_prepare/step1_3_storybooks_test.csv'
+storybooks_dataframe = pandas.read_csv(storybooks_csv_path)
+val_y = storybooks_dataframe['reading_level']
+
+# Load Predicted values from step3_2_predictions.csv
+val_predictions = pandas.read_csv('../step3_predict/step3_2_predictions.csv')
+
+accuracy = accuracy_score(val_y, val_predictions)
+print(basename(__file__), f'accuracy_score: {accuracy}')
+
+mae = mean_absolute_error(val_y, val_predictions)
+print(basename(__file__), f'accuracy_score: {mae}')
+
+# Save the results to a file for the GitHub workflow to read
+with open('Metrics_output.txt', 'w') as f:
+    f.write(f'accuracy_score: {accuracy}\n')
+    f.write(f'MAE: {mae}\n')