Add log messages

README.md

@@ -6,7 +6,7 @@ colorTo: gray
 sdk: docker
 pinned: false
 license: apache-2.0
-short_description: API endpoint for Scene understanding using Llama 3.2 Vision
+short_description: API endpoint for Scene understanding using Moondream2
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -27,18 +27,22 @@ tokenizer = AutoTokenizer.from_pretrained(model_id, revision=revision)
 
 @app.get("/")
 def read_root():
-    data = {"Voice": "Cloning", "Status": "Success"}
+    data = {"Scene": "Understanding", "Status": "Success"}
     return JSONResponse(content=data)
 
 @app.post("/generate-text/")
 async def generate_text(description: str = Form(...), file: UploadFile = File(...)):
+    print("generate_text endpoint called with description:", description)
+
     # Convert uploaded file to PIL image
     image = Image.open(file.file).convert("RGB")
+    print("Image uploaded and converted successfully")
 
     # Encode the image using the model
     enc_image = model.encode_image(image)
 
     # Answer the question using the model and tokenizer
     generated_text = model.answer_question(enc_image, description, tokenizer)
+    print("Text generated successfully")
 
     return {"generated_text": generated_text}

client.py

@@ -1,8 +1,8 @@
 import requests
 
-url = "http://127.0.0.1:8000/generate-text/"
+url = "https://lord-reso-scene-understanding.hf.space/generate-text/"
 description = "Describe this image highlighting the positions of the objects. Use simple English words."
-file_path = "assets/test.png"
+file_path = "assets/room.jpg"
 
 with open(file_path, "rb") as image_file:
     files = {"file": image_file}