Redefine cache directory based on latest standard

Dockerfile

@@ -1,8 +1,8 @@
 # Use Python as the base image
 FROM python:3.9
 
-# Set environment variable for cache
-ENV TRANSFORMERS_CACHE=/app/cache
+# Set environment variable for cache to a writable location
+ENV HF_HOME=/tmp/hf_home
 
 # Set the working directory
 WORKDIR /app
@@ -14,8 +14,8 @@ RUN pip install --no-cache-dir -r requirements.txt
 # Copy the FastAPI app
 COPY . .
 
-# Ensure the cache directory is writable
-RUN mkdir -p /app/cache
+# Ensure the cache directory is created and writable
+RUN mkdir -p /tmp/hf_home
 
 # Run FastAPI server
 CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -3,11 +3,10 @@ from fastapi import FastAPI
 from pydantic import BaseModel
 from transformers import pipeline
 
-CACHE_DIR = "/app/cache"
-
-# Ensure the cache directory is writable and exists
-os.environ["TRANSFORMERS_CACHE"] = CACHE_DIR
-os.makedirs(CACHE_DIR, exist_ok=True)  # Create the directory if it doesn't exist
+# Set the cache directory to a writable location using HF_HOME
+cache_dir = "/tmp/hf_home"
+os.environ["HF_HOME"] = cache_dir
+os.makedirs(cache_dir, exist_ok=True)
 
 # Load your Hugging Face model
 model = pipeline("text-generation", model="devops-bda/Abap")
@@ -23,7 +22,7 @@ class InputData(BaseModel):
 @app.get("/health")
 async def health_check():
     return {"status": "ok", "message": "Model is ready"}
-    
+
 # Define prediction endpoint
 @app.post("/predict")
 async def predict(data: InputData):