Create app.py

app.py

@@ -0,0 +1,41 @@
+import gradio as gr
+from transformers import pipeline
+from PIL import Image
+
+# 1) Load a pretrained image-based facial emotion model from Hugging Face
+#    The "mehdi-wasi/facial_emotion_recognition" model is an example.
+#    You can search "emotion", "FER" or "facial expression" on huggingface.co/models
+emotion_pipeline = pipeline(
+    task="image-classification",
+    model="mehdi-wasi/facial_emotion_recognition"
+)
+
+# 2) Define a function to handle image input and return predictions
+def predict_emotion(image: Image.Image):
+    """
+    Takes a PIL image, uses the emotion_pipeline to get predictions,
+    and returns them as a list of {label, score}.
+    """
+    # The pipeline will produce a list of predictions sorted by confidence
+    predictions = emotion_pipeline(image)
+
+    # We can return the raw pipeline output or we can format it
+    # e.g., only return top label or return them all
+    # Let’s return them all for clarity.
+    return predictions
+
+# 3) Build the Gradio interface
+demo = gr.Interface(
+    fn=predict_emotion,
+    inputs=gr.Image(type="pil"),  # user uploads an image (PIL format)
+    outputs="json",               # we’ll return the pipeline’s JSON predictions
+    title="Quantum Emotion Detection",
+    description="""
+    **QuantumEmotion** - A simple demo that uses a Hugging Face model to classify 
+    facial emotions. Upload an image with a face, and you'll see predicted emotions.
+    """
+)
+
+# 4) Launch the app (Gradio automatically does this on Spaces)
+if __name__ == "__main__":
+    demo.launch()