added check

README.md

@@ -11,7 +11,7 @@ pinned: false
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
-
+# Sending image file to api 
 curl -X POST "https://alexgenovese-segmentation.hf.space/segment" \
   -F "data=[{\"type\": \"image\", \"value\": null}]" \
   -F "data=@woman_with_bag.jpeg" \
@@ -19,15 +19,22 @@ curl -X POST "https://alexgenovese-segmentation.hf.space/segment" \
   -o response.json
 
 
+# Sending image file to api 
 curl -X POST "https://alexgenovese-segmentation.hf.space/segment" \
   -H "accept: application/json" \
   -H "Content-Type: multipart/form-data" \
   -F "file=@woman_with_bag.jpeg"
 
 
+# Sending image in base 64 to api
+1. base64 woman_with_bag.jpeg > image.b64
+
+2. curl -X POST "http://localhost:7860/segment" \
+-H "Content-Type: application/json" \
+-d "{\"image_base64\": \"$(cat image.b64)\"}" 
 
-curl -X POST -F "file=@woman_with_bag.jpeg" https://alexgenovese-segmentation.hf.space/api/segment
+ir 
 
-curl -X POST https://alexgenovese-segmentation.hf.space/segment \
+curl -X POST "http://localhost:7860/segment" \
 -H "Content-Type: application/json" \
 -d "{\"image_base64\": \"$(base64 woman_with_bag.jpeg)\"}"

app.py

@@ -1,29 +1,43 @@
-from fastapi import FastAPI, File, UploadFile
+from fastapi import FastAPI, File, UploadFile, HTTPException
 from transformers import SamModel, SamProcessor
 import torch
 from PIL import Image
 import numpy as np
 import io
 import base64
+import logging
 
 # Inizializza l'app FastAPI
 app = FastAPI()
 
+# Configura il logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
 # Carica il modello e il processore SAM
-model = SamModel.from_pretrained("facebook/sam-vit-base")
-processor = SamProcessor.from_pretrained("facebook/sam-vit-base")
-model.to("cpu")  # Usa CPU per il free tier
+try:
+    logger.info("Caricamento del modello SAM...")
+    model = SamModel.from_pretrained("facebook/sam-vit-base")
+    processor = SamProcessor.from_pretrained("facebook/sam-vit-base")
+    model.to("cpu")  # Usa CPU per il free tier
+    logger.info("Modello caricato con successo.")
+except Exception as e:
+    logger.error(f"Errore nel caricamento del modello: {str(e)}")
+    raise RuntimeError(f"Errore nel caricamento del modello: {str(e)}")
 
 # Funzione per segmentare l'immagine
 def segment_image(image: Image.Image):
     # Prepara l'input per SAM
+    logger.info("Preparazione dell'immagine per l'inferenza...")
     inputs = processor(image, return_tensors="pt").to("cpu")
     
     # Inferenza
+    logger.info("Esecuzione dell'inferenza...")
     with torch.no_grad():
         outputs = model(**inputs, multimask_output=False)
     
     # Post-processa la maschera
+    logger.info("Post-processing della maschera...")
     mask = processor.image_processor.post_process_masks(
         outputs.pred_masks, inputs["original_sizes"], inputs["reshaped_input_sizes"]
     )[0][0].cpu().numpy()
@@ -44,18 +58,21 @@ def segment_image(image: Image.Image):
 # Endpoint API
 @app.post("/segment")
 async def segment_endpoint(file: UploadFile = File(...)):
-    # Leggi l'immagine caricata
-    image_data = await file.read()
-    image = Image.open(io.BytesIO(image_data)).convert("RGB")
-    
-    # Segmenta l'immagine
-    mask_base64, annotations = segment_image(image)
-    
-    # Restituisci la risposta
-    return {
-        "mask": f"data:image/png;base64,{mask_base64}",
-        "annotations": annotations
-    }
+    try:
+        logger.info("Ricezione del file...")
+        image_data = await file.read()
+        image = Image.open(io.BytesIO(image_data)).convert("RGB")
+        
+        logger.info("Segmentazione dell'immagine...")
+        mask_base64, annotations = segment_image(image)
+        
+        return {
+            "mask": f"data:image/png;base64,{mask_base64}",
+            "annotations": annotations
+        }
+    except Exception as e:
+        logger.error(f"Errore nell'endpoint: {str(e)}")
+        raise HTTPException(status_code=500, detail=f"Errore nell'elaborazione: {str(e)}")
 
 # Per compatibilità con Hugging Face Spaces (Uvicorn viene gestito automaticamente)
 if __name__ == "__main__":

requirements.txt

@@ -1,6 +1,7 @@
+fastapi
+uvicorn
 torch
 torchvision
 transformers
-gradio
 pillow
 numpy