Add Streamlit app and requirements

app.py

@@ -0,0 +1,39 @@
+import streamlit as st
+import requests
+from PIL import Image
+import io
+
+# FastAPI backend URL for prediction (replace this with your actual FastAPI URL)
+FASTAPI_BACKEND_URL = "https://your-fastapi-model-url.a.run.app/predict"
+
+# Function to send the image to FastAPI and get the prediction
+def get_prediction(image):
+    # Convert image to bytes
+    img_byte_array = io.BytesIO()
+    image.save(img_byte_array, format="JPEG")
+    img_byte_array = img_byte_array.getvalue()
+
+    # Send the image to FastAPI using a POST request
+    response = requests.post(FASTAPI_BACKEND_URL, files={"file": img_byte_array})
+
+    if response.status_code == 200:
+        prediction = response.json()
+        return prediction
+    else:
+        return {"error": "Prediction failed"}
+
+# Upload image through Streamlit
+uploaded_image = st.file_uploader("Upload an image of a tree", type=["jpg", "png"])
+
+if uploaded_image:
+    # Display the image
+    image = Image.open(uploaded_image)
+    st.image(image, caption="Uploaded Tree Image")
+
+    # Make prediction on the image
+    if st.button("Predict Decoration"):
+        prediction = get_prediction(image)
+        if 'error' not in prediction:
+            st.write(f"Prediction: {prediction['prediction']}")
+        else:
+            st.write("Error occurred during prediction.")

requirements.txt

@@ -0,0 +1,3 @@
+streamlit
+requests
+pillow