Reduce runtime

.github/workflows/build.yml

@@ -18,11 +18,11 @@ jobs:
             CMD_BUILD: python -m build --wheel
 
     steps:
-      - uses: actions/checkout@v4
-      - name: Set up Python 3.9
-        uses: actions/setup-python@v5
+      - uses: actions/checkout@v5
+      - name: Set up Python 3.11
+        uses: actions/setup-python@v6
         with:
-          python-version: "3.9"
+          python-version: "3.11"
 
       - name: Install dependencies
         run: |
@@ -33,7 +33,7 @@ jobs:
         run: ${{matrix.CMD_BUILD}}
 
       - name: Upload Python wheel
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v7
         with:
           name: Python wheel
           path: ${{github.workspace}}/dist/raidionicsval-*.whl
@@ -42,12 +42,12 @@ jobs:
   setup-test-data:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
         with:
-          python-version: 3.9
+          python-version: 3.11
 
       - name: Download test resources
         working-directory: tests
@@ -56,7 +56,7 @@ jobs:
             python -c "from download_resources import download_resources; download_resources('../test_data')"
 
       - name: Upload test resources
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v7
         with:
           name: test-resources
           path: ./test_data
@@ -106,14 +106,6 @@ jobs:
             python-version: "3.12"
           - os: windows-2025
             python-version: "3.13"
-          - os: macos-13
-            python-version: "3.9"
-          - os: macos-13
-            python-version: "3.10"
-          - os: macos-13
-            python-version: "3.11"
-          - os: macos-13
-            python-version: "3.12"
           - os: macos-14
             python-version: "3.10"
           - os: macos-14
@@ -130,15 +122,14 @@ jobs:
             python-version: "3.12"
           - os: macos-15
             python-version: "3.13"
-
     steps:
         - name: Set up Python ${{ matrix.python-version }}
-          uses: actions/setup-python@v5
+          uses: actions/setup-python@v6
           with:
             python-version: ${{ matrix.python-version }}
 
         - name: Download artifact
-          uses: actions/download-artifact@v4
+          uses: actions/download-artifact@v8
           with:
             name: "Python wheel"
 
@@ -152,22 +143,36 @@ jobs:
           run: raidionicsval --help
 
         - name: Clone repo
-          uses: actions/checkout@v4
+          uses: actions/checkout@v5
 
         - name: Download test resources
-          uses: actions/download-artifact@v4
+          uses: actions/download-artifact@v8
           with:
             name: test-resources
             path: ./tests/unit_tests_results_dir
 
         - name: Integration tests
+          env: 
+              MPLBACKEND: Agg
           run: |
               pip install pytest pytest-cov pytest-timeout requests
-              pytest -vvv --cov=raidionicsval ${{github.workspace}}/tests/generic_tests --cov-report=xml --timeout=1000 --log-cli-level=DEBUG
+              pytest -vvv --cov=raidionicsval ${{github.workspace}}/tests/generic_tests --cov-report=xml --timeout=1000 --log-cli-level=INFO
+
+        - name: Upload integration test results
+          if: failure()
+          uses: actions/upload-artifact@v7
+          with:
+            name: integration-test-results-${{ matrix.os }}-py${{ matrix.python-version }}
+            path: |
+                tests/unit_tests_results_dir/**
+                !tests/unit_tests_results_dir/**/Inputs/**
+                !tests/unit_tests_results_dir/**/Predictions/**
+            if-no-files-found: ignore
+            retention-days: 1
 
         - name: Upload coverage to Codecov
-          if: ${{ matrix.os == 'ubuntu-22.04' && matrix.python-version == '3.9' }}
-          uses: codecov/codecov-action@v4
+          if: ${{ matrix.os == 'ubuntu-22.04' && matrix.python-version == '3.11' }}
+          uses: codecov/codecov-action@v6
           with:
             token: ${{ secrets.CODECOV_TOKEN  }}
             slug: dbouget/validation_metrics_computation

main.py

@@ -42,4 +42,6 @@ def main(argv):
 
 
 if __name__ == "__main__":
+    import matplotlib
+    matplotlib.use("Agg")
     main(sys.argv[1:])

raidionicsval/Utils/PatientMetricsStructure.py

@@ -106,6 +106,87 @@ def init_from_file(self, study_folder: str):
         else:
             self.__init_from_file_classification(study_folder=study_folder)
 
+    def init_from_db(self, conn):
+        if self.objective == "segmentation":
+            self.__init_from_db_segmentation(conn)
+        else:
+            # @TODO. Add database-backed loading for classification objective
+            pass
+
+    def __init_from_db_segmentation(self, conn):
+        """
+        Loads segmentation metrics for this patient from SQLite.
+        Fetches per-class results first, then the macro-averaged total_results rows.
+        Returns early if no data exists yet (i.e. first run).
+        """
+        # Load per-class metrics using ClassMetrics.init_from_df
+        for c in list(self._class_metrics.keys()):
+            table_name = f"class_{c}"
+            try:
+                rows_df = pd.read_sql_query(
+                    f"SELECT * FROM [{table_name}] WHERE [Patient] = ? AND [Fold] = ?",
+                    conn,
+                    params=(str(self._patient_id), self._fold_number)
+                )
+                self._class_metrics[c].init_from_df(rows_df)
+            except Exception:
+                return # Table does not exist yet on first run
+
+        # Load macro-averaged results from total_results
+        rows_df = pd.read_sql_query(
+            "SELECT * FROM [total_results] WHERE [Patient] = ? AND [Fold] = ?",
+            conn,
+            params=(str(self._patient_id), self._fold_number)
+        )
+
+        if len(rows_df) == 0:
+            return
+
+        def to_float(v):
+            # Convert to float since SQLite returns numeric values as strings
+            try:
+                return float(v) if v is not None else None
+            except (ValueError, TypeError):
+                return None
+
+        self._patientwise_metrics = []
+        self._pixelwise_metrics = []
+        self._objectwise_metrics = []
+        self._extra_metrics = []
+
+        upper_idx = SharedResources.getInstance().upper_default_metrics_index
+        extra_metric_names = []
+        for m in SharedResources.getInstance().validation_metric_names:
+            extra_metric_names.extend([f'PiW {m}', f'OW {m}'])
+
+        for thr in rows_df["Threshold"].values:
+            thr_results = rows_df.loc[rows_df["Threshold"] == thr].values[0]
+            thr_val = float(thr_results[2])
+            pixelwise_values = [to_float(x) for x in thr_results[3:7]]
+            patientwise_values = [to_float(x) for x in thr_results[7:10]]
+            objectwise_values = [to_float(x) for x in thr_results[10:upper_idx]]
+            extra_values = [to_float(x) for x in thr_results[upper_idx:]]
+
+            # Pad missing extra metrics with None to match expected length
+            while len(extra_values) < 2 * len(SharedResources.getInstance().validation_metric_names):
+                extra_values.append(None)
+
+            extra_values_description = list(rows_df.columns[upper_idx:])
+            for name in extra_metric_names:
+                if name not in extra_values_description:
+                    extra_values_description.append(name)
+
+            self._pixelwise_metrics.append([thr_val] + pixelwise_values)
+            self._patientwise_metrics.append([thr_val] + patientwise_values)
+            self._objectwise_metrics.append([thr_val] + objectwise_values)
+
+            extra_values_cat = [[x, y] for x, y in zip(extra_values_description, extra_values)]
+            if extra_values_cat:
+                self._extra_metrics.append([thr_val] + extra_values_cat)
+
+        if len(self._extra_metrics) == 0:
+            self._extra_metrics = None
+
     def __init_from_file_segmentation(self, study_folder: str):
         all_scores_filename = os.path.join(study_folder, 'all_dice_scores.csv')
 
@@ -313,7 +394,7 @@ def set_results(self, results):
         self._objectwise_metrics = []
         for index in range(len(results)):
             thr_results = results[index][0]
-            thr_val = thr_results[2]
+            thr_val = float(thr_results[2])
             pixelwise_values = thr_results[3:7]
             patientwise_values = thr_results[7:10]
             objectwise_values = thr_results[10:SharedResources.getInstance().upper_default_metrics_index]
@@ -358,6 +439,59 @@ def init_from_file(self, scores_filename: str) -> None:
         if len(self._extra_metrics) == 0:
             self._extra_metrics = None
 
+    def init_from_df(self, rows_df: pd.DataFrame) -> None:
+        """
+        Initializes class-level metrics from a pre-fetched DataFrame of per-threshold rows.
+        """
+        if len(rows_df) == 0:
+            return
+
+        upper_idx = SharedResources.getInstance().upper_default_metrics_index
+
+        def to_float(v):
+            # Convert to float since SQLite returns numeric values as strings
+            try:
+                return float(v) if v is not None else None
+            except (ValueError, TypeError):
+                return None
+
+        self._patientwise_metrics = []
+        self._pixelwise_metrics = []
+        self._objectwise_metrics = []
+        self._extra_metrics = []
+
+        extra_metric_names = []
+        for m in SharedResources.getInstance().validation_metric_names:
+            extra_metric_names.extend([f'PiW {m}', f'OW {m}'])
+
+        for thr in np.unique(rows_df["Threshold"].values):
+            thr_results = rows_df.loc[rows_df["Threshold"] == thr].values[0]
+            thr_val = float(thr_results[2])
+            pixelwise_values = [to_float(x) for x in thr_results[3:7]]
+            patientwise_values = [to_float(x) for x in thr_results[7:10]]
+            objectwise_values = [to_float(x) for x in thr_results[10:upper_idx]]
+            extra_values = [to_float(x) for x in thr_results[upper_idx:]]
+
+            # Pad missing extra metrics with NaN to match expected length
+            while len(extra_values) < 2 * len(SharedResources.getInstance().validation_metric_names):
+                extra_values.append(float('nan'))
+
+            extra_values_description = list(rows_df.columns[upper_idx:])
+            for name in extra_metric_names:
+                if name not in extra_values_description:
+                    extra_values_description.append(name)
+
+            self._pixelwise_metrics.append([thr_val] + pixelwise_values)
+            self._patientwise_metrics.append([thr_val] + patientwise_values)
+            self._objectwise_metrics.append([thr_val] + objectwise_values)
+
+            extra_values_cat = [[x, y] for x, y in zip(extra_values_description, extra_values)]
+            if extra_values_cat:
+                self._extra_metrics.append([thr_val] + extra_values_cat)
+
+        if len(self._extra_metrics) == 0:
+            self._extra_metrics = None
+
     def is_complete(self):
         """