Update app/sentiment.py

app/sentiment.py

@@ -1,32 +1,21 @@
 """
-Safe lazy-loading sentiment pipeline that works in Hugging Face Spaces (no /.cache error).
+Sentiment analysis module using Hugging Face Inference API to avoid local model downloads.
 """
-
 import os
 import hashlib
 import logging
 from functools import lru_cache
+import httpx
 
-# Redirect the HF model cache to a writable directory
-os.environ["TRANSFORMERS_CACHE"] = "/tmp/huggingface"
-os.makedirs("/tmp/huggingface", exist_ok=True)
-
-from transformers import pipeline
+# Environment variables (set HF_API_TOKEN in your Space's Settings)
+HF_API_TOKEN = os.getenv("HF_API_TOKEN", "")
+API_URL = "https://api-inference.huggingface.co/models/distilbert-base-uncased-finetuned-sst-2-english"
+HEADERS = {"Authorization": f"Bearer {HF_API_TOKEN}"}
 
+# In-memory cache for latest sentiment
 class SentimentCache:
     latest_id: int = 0
     latest_result: dict = {}
-    _pipeline = None  # Lazy init
-
-    @classmethod
-    def _get_pipeline(cls):
-        if cls._pipeline is None:
-            logging.info("🔄 Loading sentiment model…")
-            cls._pipeline = pipeline(
-                "sentiment-analysis",
-                model="distilbert-base-uncased-finetuned-sst-2-english"
-            )
-        return cls._pipeline
 
     @classmethod
     def _hash(cls, text: str) -> str:
@@ -35,16 +24,26 @@ class SentimentCache:
     @classmethod
     @lru_cache(maxsize=128)
     def _analyze(cls, text: str):
-        pipe = cls._get_pipeline()
-        return pipe(text)[0]
+        try:
+            response = httpx.post(API_URL, headers=HEADERS, json={"inputs": text}, timeout=20)
+            response.raise_for_status()
+            data = response.json()
+            # Expecting list of {label, score}
+            if isinstance(data, list) and data:
+                return data[0]
+            raise ValueError("Unexpected response format: %s" % data)
+        except Exception as e:
+            logging.error("❌ Sentiment API error: %s", e)
+            return {"label": "ERROR", "score": 0.0}
 
     @classmethod
     def compute(cls, text: str):
+        """Trigger sentiment inference via API and update latest result."""
         res = cls._analyze(text)
         cls.latest_id += 1
         cls.latest_result = {
             "text": text,
-            "label": res["label"],
-            "score": round(res["score"], 4)
+            "label": res.get("label"),
+            "score": round(res.get("score", 0.0), 4)
         }
         logging.info("✅ Sentiment computed: %s", cls.latest_result)