Update app.py

added user feedback

app.py

@@ -3,6 +3,9 @@ from torchvision import models, transforms
 from PIL import Image
 import gradio as gr
 import matplotlib.pyplot as plt
+import pandas as pd
+import os
+import hashlib
 from huggingface_hub import hf_hub_download
 
 # Modell laden vom Hugging Face Model Hub
@@ -28,6 +31,15 @@ transform = transforms.Compose([
     transforms.ToTensor()
 ])
 
+# Feedback-File
+FEEDBACK_FILE = "user_feedback.csv"
+
+# Hilfsfunktion für Hash des Bildes
+def get_image_hash(image):
+    img_bytes = image.tobytes()
+    return hashlib.md5(img_bytes).hexdigest()
+
+# Plot-Funktion
 def plot_probabilities(probabilities, labels):
     probs = probabilities.cpu().numpy().flatten()
     fig, ax = plt.subplots(figsize=(8, 4))
@@ -39,12 +51,13 @@ def plot_probabilities(probabilities, labels):
     plt.tight_layout()
     return fig
 
+# Prediction-Funktion
 def predict_emotion(image):
     image = image.convert("RGB")
-    image = transform(image).unsqueeze(0).to(device)
+    transformed_image = transform(image).unsqueeze(0).to(device)
 
     with torch.no_grad():
-        outputs = model(image)
+        outputs = model(transformed_image)
         probs = torch.softmax(outputs, dim=1)
 
     # Top 3 Predictions
@@ -64,21 +77,52 @@ def predict_emotion(image):
     # Bar Chart
     fig = plot_probabilities(probs, labels)
 
-    # Ausgabe
-    return prediction, f"Confidence: {confidence.item()*100:.2f}%\n{prediction_status}", top3, fig
+    # Bild-Hash für spätere Zuordnung
+    img_hash = get_image_hash(image)
+
+    return prediction, f"Confidence: {confidence.item()*100:.2f}%\n{prediction_status}", top3, fig, img_hash
+
+# Funktion um Feedback zu speichern
+def save_feedback(img_hash, model_prediction, user_feedback, confidence):
+    data = {
+        "image_hash": [img_hash],
+        "model_prediction": [model_prediction],
+        "user_feedback": [user_feedback],
+        "confidence": [confidence]
+    }
+    df_new = pd.DataFrame(data)
+    if os.path.exists(FEEDBACK_FILE):
+        df_existing = pd.read_csv(FEEDBACK_FILE)
+        df_existing = pd.concat([df_existing, df_new], ignore_index=True)
+        df_existing.to_csv(FEEDBACK_FILE, index=False)
+    else:
+        df_new.to_csv(FEEDBACK_FILE, index=False)
+    return "✅ Vielen Dank für dein Feedback!"
+
+# Kombinierte Funktion
+def full_pipeline(image, user_feedback):
+    prediction, confidence_text, top3, fig, img_hash = predict_emotion(image)
+    feedback_message = save_feedback(img_hash, prediction, user_feedback, confidence_text.split("\n")[0])
+    return prediction, confidence_text, top3, fig, feedback_message
 
 # Gradio Interface
-interface = gr.Interface(
-    fn=predict_emotion,
-    inputs=gr.Image(type="pil"),
-    outputs=[
-        gr.Textbox(label="Vorhergesagte Emotion"),
-        gr.Textbox(label="Confidence + Einschätzung"),
-        gr.Dataframe(headers=["Emotion", "Wahrscheinlichkeit (%)"], label="Top 3 Emotionen"),
-        gr.Plot(label="Verteilung der Emotionen")
-    ],
-    title="Emotion Recognition App",
-    description="Lade ein Bild hoch und erkenne die Emotion. Zeigt auch die Top 3 Emotionen und alle Wahrscheinlichkeiten als Balkendiagramm."
-)
+with gr.Blocks() as interface:
+    with gr.Row():
+        with gr.Column():
+            image_input = gr.Image(type="pil", label="Lade dein Bild hoch")
+            feedback_input = gr.Dropdown(choices=labels, label="Dein Feedback: Was ist die richtige Emotion?")
+            submit_btn = gr.Button("Absenden")
+        with gr.Column():
+            prediction_output = gr.Textbox(label="Vorhergesagte Emotion")
+            confidence_output = gr.Textbox(label="Confidence + Einschätzung")
+            top3_output = gr.Dataframe(headers=["Emotion", "Wahrscheinlichkeit (%)"], label="Top 3 Emotionen")
+            plot_output = gr.Plot(label="Verteilung der Emotionen")
+            feedback_message_output = gr.Textbox(label="Feedback-Status")
+
+    submit_btn.click(
+        fn=full_pipeline,
+        inputs=[image_input, feedback_input],
+        outputs=[prediction_output, confidence_output, top3_output, plot_output, feedback_message_output]
+    )
 
 interface.launch()