fixed errors

app.py

@@ -1,21 +1,20 @@
-from transformers import BlipProcessor, BlipForConditionalGeneration  # For image captioning using BLIP model
+from transformers import BlipProcessor, BlipForConditionalGeneration
 import io
 from PIL import Image
 from fastapi import FastAPI, File, UploadFile, HTTPException
 from fastapi.responses import JSONResponse
-import uvicorn                                               
-import torch                                                          # For running deep learning models (PyTorch backend)
+import uvicorn
+import torch
 
-# Load BLIP (Bootstrapping Language-Image Pretraining) model
+# ── Load BLIP Model ────────────────────────────────────────────────────────
 processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
 model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
 
-
-
-
-def image_to_speech(image):
+def image_to_speech(image: Image.Image) -> str:
     # Step 1: Generate a more detailed caption
+    # The processor handles the conversion from PIL Image to PyTorch tensors
     inputs = processor(image, return_tensors="pt")
+    
     out = model.generate(
         **inputs,
         max_length=90,          # Allow longer, more detailed sentences
@@ -25,55 +24,48 @@ def image_to_speech(image):
         early_stopping=True
     )
     caption = processor.decode(out[0], skip_special_tokens=True)
+    return caption # Removed the trailing comma
 
-    return caption,
-
-# ── FastAPI app ────────────────────────────────────────────────────────────────
+# ── FastAPI app ────────────────────────────────────────────────────────────
 app = FastAPI(
     title="VocalEyes API",
-    description="Converts an uploaded image into a short scene description via RelTR + T5.",
+    description="Converts an uploaded image into a short scene description via BLIP.",
     version="1.0.0",
 )
 
-
 @app.get("/")
 def root():
     return {"status": "ok", "message": "VocalEyes API is running. POST an image to /predict"}
 
-
 @app.post("/predict")
 async def predict(file: UploadFile = File(...)):
-    # ── Validate content type ──────────────────────────────────────────────────
+    # ── Validate content type ──────────────────────────────────────────────
     if file.content_type not in ("image/jpeg", "image/png", "image/webp", "image/bmp"):
         raise HTTPException(
             status_code=415,
             detail=f"Unsupported file type '{file.content_type}'. Send JPEG, PNG, WEBP, or BMP.",
         )
-
-    # ── Read & preprocess ──────────────────────────────────────────────────────
+        
+    # ── Read & preprocess ──────────────────────────────────────────────────
     try:
         raw = await file.read()
         image = Image.open(io.BytesIO(raw)).convert("RGB")
     except Exception as e:
         raise HTTPException(status_code=400, detail=f"Could not read image: {e}")
 
-    img_tensor = transform(image).unsqueeze(0)   # (1, 3, H, W)
-
-    # ── Run pipeline ───────────────────────────────────────────────────────────
+    # ── Run pipeline ───────────────────────────────────────────────────────
     try:
-        result = image_to_speech(img_tensor)
+        # Pass the PIL Image directly to our BLIP function
+        caption = image_to_speech(image)
     except Exception as e:
         raise HTTPException(status_code=500, detail=f"Pipeline error: {e}")
 
+    # Return the clean description
     return JSONResponse({
-        "description": result["generated_text"],
-        "triplets": [
-            {"subject": s, "relation": r, "object": o}
-            for s, r, o in result["triplets"]
-        ],
+        "description": caption
     })
 
-
-# ── Entry point ────────────────────────────────────────────────────────────────
+# ── Entry point ──────────────────────────────────────────────────���─────────
 if __name__ == "__main__":
+    # Make sure this file is saved as app.py if you are passing "app:app"
     uvicorn.run("app:app", host="0.0.0.0", port=7860)