Add initial FastAPI application with Docker support and Redis caching

Dockerfile

@@ -0,0 +1,10 @@
+FROM python:3.9
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,46 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+from sentence_transformers import SentenceTransformer
+import torch
+import redis
+import json
+import hashlib
+import os
+
+app = FastAPI()
+model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
+
+# Redis connection using Hugging Face storage
+REDIS_HOST = os.getenv("REDIS_HOST", "localhost")
+redis_client = redis.Redis(host=REDIS_HOST, port=6379, db=0, decode_responses=True)
+
+class EmbedRequest(BaseModel):
+    text: str
+
+def get_cache_key(text: str) -> str:
+    return f"embed:{hashlib.md5(text.encode()).hexdigest()}"
+
+@app.get("/")
+async def root():
+    return {"message": "Embedding API is running! Use /embed to generate embeddings."}
+
+@app.post("/embed")
+async def embed(request: EmbedRequest):
+    text = request.text.strip()
+    if not text:
+        raise HTTPException(status_code=400, detail="Text cannot be empty")
+    
+    cache_key = get_cache_key(text)
+    cached_result = redis_client.get(cache_key)
+    if cached_result:
+        return json.loads(cached_result)
+
+    embedding = model.encode([text], convert_to_tensor=True).cpu().tolist()[0]
+    redis_client.setex(cache_key, 86400, json.dumps({"embedding": embedding}))
+    
+    return {"embedding": embedding}
+
+@app.get("/health")
+async def health_check():
+    redis_status = "ok" if redis_client.ping() else "down"
+    return {"status": "ok", "gpu": torch.cuda.is_available(), "redis": redis_status}

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+torch
+sentence-transformers
+redis