Enhance README and app.py: Add new analysis tasks and improve MusicXML handling

README.md

@@ -25,8 +25,13 @@ A Gradio web interface for [AnalysisGNN](https://github.com/manoskary/analysisGN
   - Harmonic Analysis (Chord Quality, Root, Bass, Inversion)
   - Roman Numeral Analysis
   - Phrase & Section Segmentation
+  - Harmonic Rhythm
+  - Pitch-Class Set Groupings
+  - Non-Chord Tone (TPC-in-label / NCT) Detection
+  - Note Degree Labeling
 - 📈 **Results Table**: View analysis results in an interactive table
 - 💾 **Export Results**: Download analysis results as CSV
+- 🧾 **Parsed Score Download**: Grab the normalized MusicXML that is produced after parsing with Partitura
 
 ## Quick Start
 

app.py

@@ -23,7 +23,11 @@ warnings.filterwarnings('ignore')
 # Import partitura and AnalysisGNN
 import partitura as pt
 from analysisgnn.models.analysis import ContinualAnalysisGNN
-from analysisgnn.utils.chord_representations import available_representations
+from analysisgnn.utils.chord_representations import available_representations, NoteDegree49
+
+# Ensure additional representations are available for decoding
+if "note_degree" not in available_representations and NoteDegree49 is not None:
+    available_representations["note_degree"] = NoteDegree49
 
 # Global model variable
 MODEL = None
@@ -101,6 +105,41 @@ def load_model() -> ContinualAnalysisGNN:
     return MODEL
 
 
+def resolve_musicxml_path(musicxml_file) -> Optional[str]:
+    """Return a filesystem path for the uploaded MusicXML file."""
+    if musicxml_file is None:
+        return None
+    if isinstance(musicxml_file, (str, os.PathLike)):
+        return str(musicxml_file)
+    if isinstance(musicxml_file, dict) and "name" in musicxml_file:
+        return musicxml_file["name"]
+    file_path = getattr(musicxml_file, "name", None)
+    if file_path:
+        return file_path
+    return None
+
+
+def save_parsed_musicxml(score: pt.score.Score, original_path: Optional[str]) -> Optional[str]:
+    """
+    Persist the parsed/normalized score to a temporary MusicXML file.
+    
+    Returns the path to the saved file or None if saving fails.
+    """
+    try:
+        suffix = ".musicxml"
+        if original_path:
+            original_suffix = Path(original_path).suffix.lower()
+            if original_suffix in {".xml", ".musicxml"}:
+                suffix = original_suffix
+        fd, tmp_path = tempfile.mkstemp(suffix=suffix)
+        os.close(fd)
+        pt.save_musicxml(score, tmp_path)
+        return tmp_path
+    except Exception as exc:
+        print(f"Warning: Could not save parsed MusicXML: {exc}")
+        return None
+
+
 def render_score_to_image(score: pt.score.Score, output_path: str) -> Optional[str]:
     """
     Render score to image using partitura.
@@ -140,7 +179,7 @@ def render_score_to_image(score: pt.score.Score, output_path: str) -> Optional[s
             except (ImportError, Exception) as e:
                 # If conversion fails, return the PDF path
                 print(f"PDF to PNG conversion failed: {e}")
-                return pdf_path
+                print("Score rendered as PDF but could not be converted to PNG for visualization.")
     except Exception as e:
         print(f"Error rendering score to PDF: {e}")
     
@@ -231,7 +270,7 @@ def predict_analysis(
 def process_musicxml(
     musicxml_file,
     selected_tasks: list
-) -> Tuple[Optional[str], Optional[pd.DataFrame], str]:
+) -> Tuple[Optional[str], Optional[pd.DataFrame], Optional[str], str]:
     """
     Process a MusicXML file and return visualization and analysis results.
     
@@ -245,15 +284,19 @@ def process_musicxml(
     Returns
     -------
     tuple
-        (image_path, dataframe, status_message)
+        (image_path, dataframe, parsed_musicxml_path, status_message)
     """
     if musicxml_file is None:
-        return None, None, "Please upload a MusicXML file."
+        return None, None, None, "Please upload a MusicXML file."
     
     if not selected_tasks:
-        return None, None, "Please select at least one analysis task."
+        return None, None, None, "Please select at least one analysis task."
     
     try:
+        score_path = resolve_musicxml_path(musicxml_file)
+        if score_path is None or not os.path.exists(score_path):
+            return None, None, None, "Could not locate the uploaded MusicXML file."
+        
         # Load the model
         status_msg = "Loading model..."
         print(status_msg)
@@ -262,7 +305,9 @@ def process_musicxml(
         # Load the score
         status_msg = "Loading score..."
         print(status_msg)
-        score = pt.load_musicxml(musicxml_file.name)
+        score = pt.load_musicxml(score_path)
+        
+        parsed_score_path = save_parsed_musicxml(score, score_path)
         
         # Render score to image
         status_msg = "Rendering score..."
@@ -287,6 +332,14 @@ def process_musicxml(
             if 'note_id' not in df.columns:
                 df.insert(0, 'note_id', range(len(df)))
             
+            # Convert tpc_in_label logits into NCT-friendly labels
+            if 'tpc_in_label' in df.columns:
+                df['tpc_in_label'] = np.where(
+                    df['tpc_in_label'].astype(int) == 0,
+                    "NCT",
+                    "Chord Tone"
+                )
+            
             # Reorder columns to have timing info first, then predictions, then confidence
             timing_cols = [col for col in ['note_id', 'onset_beat', 'measure'] if col in df.columns]
             confidence_cols = [col for col in df.columns if col.endswith('_confidence')]
@@ -302,17 +355,32 @@ def process_musicxml(
             
             df = df[ordered_cols]
             
+            # Apply user-friendly column names
+            rename_map = {}
+            for key, label in DISPLAY_NAME_OVERRIDES.items():
+                if key in df.columns:
+                    rename_map[key] = label
+                conf_key = f"{key}_confidence"
+                if conf_key in df.columns:
+                    rename_map[conf_key] = f"{label} Confidence"
+            if rename_map:
+                df = df.rename(columns=rename_map)
+            
             status_msg = f"✓ Analysis complete! Analyzed {len(df)} notes with {len(selected_tasks)} task(s)."
+            if parsed_score_path:
+                status_msg += " Parsed MusicXML ready for download."
         else:
             df = pd.DataFrame()
             status_msg = "⚠ Analysis returned no predictions."
+            if parsed_score_path:
+                status_msg += " Parsed MusicXML ready for download."
         
-        return rendered_path, df, status_msg
+        return rendered_path, df, parsed_score_path, status_msg
         
     except Exception as e:
         error_msg = f"Error processing file: {str(e)}\n\n{traceback.format_exc()}"
         print(error_msg)
-        return None, None, error_msg
+        return None, None, None, error_msg
 
 
 # Define available tasks
@@ -329,6 +397,15 @@ AVAILABLE_TASKS = {
     "romanNumeral": "Roman Numeral Analysis",
     "phrase": "Phrase Segmentation",
     "section": "Section Detection",
+    "hrhythm": "Harmonic Rhythm",
+    "pcset": "Pitch-Class Set",
+    "tpc_in_label": "Non-Chord Tone (NCT)",
+    "note_degree": "Note Degree",
+}
+
+DISPLAY_NAME_OVERRIDES = {
+    "tpc_in_label": "NCT",
+    "note_degree": "Note Degree",
 }
 
 # Create Gradio interface
@@ -343,6 +420,8 @@ with gr.Blocks(title="AnalysisGNN Music Analysis", theme=gr.themes.Soft()) as de
     - Key Analysis (Local & Tonalized)
     - Harmonic Analysis (Chords, Inversions, Roman Numerals)
     - Phrase & Section Segmentation
+    - Non-Chord Tone Detection (TPC-in-label / NCT)
+    - Note Degree Labeling
     
     **Model:** Pre-trained AnalysisGNN from [manoskary/analysisGNN](https://github.com/manoskary/analysisGNN)
     """)
@@ -378,14 +457,18 @@ with gr.Blocks(title="AnalysisGNN Music Analysis", theme=gr.themes.Soft()) as de
                 interactive=False
             )
     
-    with gr.Row():
-        with gr.Column():
-            # Score visualization
-            gr.Markdown("### 🎼 Score Visualization")
-            image_output = gr.Image(
-                label="Rendered Score",
-                type="filepath"
-            )
+with gr.Row():
+    with gr.Column():
+        # Score visualization
+        gr.Markdown("### 🎼 Score Visualization")
+        image_output = gr.Image(
+            label="Rendered Score",
+            type="filepath"
+        )
+        parsed_score_output = gr.File(
+            label="Parsed MusicXML (Download)",
+            interactive=False
+        )
     
     with gr.Row():
         with gr.Column():
@@ -451,7 +534,7 @@ with gr.Blocks(title="AnalysisGNN Music Analysis", theme=gr.themes.Soft()) as de
     analyze_btn.click(
         fn=analyze_wrapper,
         inputs=[file_input, task_selector],
-        outputs=[image_output, table_output, status_output]
+        outputs=[image_output, table_output, parsed_score_output, status_output]
     )
     
     example_btn.click(