Add application file

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.11
+
+WORKDIR /code
+
+COPY requirements.txt .
+
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README copy.md

@@ -0,0 +1,11 @@
+---
+title: VisionMate API
+emoji: 🧠
+colorFrom: indigo
+colorTo: pink
+sdk: docker
+app_file: app.py
+pinned: false
+---
+# VisionMate Backend
+Runs a FastAPI service using BLIP model (Salesforce).

app.py

@@ -0,0 +1,56 @@
+from fastapi import FastAPI, File, UploadFile
+from fastapi.middleware.cors import CORSMiddleware
+from PIL import Image
+from transformers import GitProcessor, AutoModelForCausalLM
+import torch
+import io
+
+
+
+app = FastAPI()
+
+# Enable CORS
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Allow all origins
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+
+# Load GIT-base model and processor
+print("🚀 Loading microsoft/git-base-coco model...")
+processor = GitProcessor.from_pretrained("microsoft/git-base-coco/")
+model = AutoModelForCausalLM.from_pretrained("microsoft/git-base-coco", torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32)
+print("✅ Model loaded!")
+
+@app.get("/")
+def read_root():
+    return {"message": "VisionMate API is running!"}
+
+@app.post("/caption/")
+async def generate_caption(file: UploadFile = File(...)):
+    print("📥 Received image upload request")
+    
+    # Read and process image
+    image = Image.open(io.BytesIO(await file.read())).convert("RGB")
+    print("🖼️ Image processed")
+
+    # Provide a better prompt to guide caption generation
+    prompt = "a photo of"
+    inputs = processor(images=image, text=prompt, return_tensors="pt")
+
+    print("🤖 Generating caption...")
+    output_ids = model.generate(**inputs, max_new_tokens=50)  # increased length
+    caption = processor.batch_decode(output_ids, skip_special_tokens=True)[0]
+
+    print("📝 Caption generated:", caption)
+    return {"caption": caption}
+
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=7860)
+

requirements.txt

@@ -0,0 +1,6 @@
+transformers
+torch
+fastapi
+uvicorn
+pillow
+python-multipart