Updated Dockerfile for Hugging Face Spaces deployment

Dockerfile

@@ -47,7 +47,7 @@ ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
 # Set environment variable for NeMo cache directory
 ENV NEMO_NLP_TMP=/app/.cache
 
-# Create cache directory
+# Create cache directory and set permissions
 RUN mkdir -p /app/.cache && chmod -R 777 /app/.cache
 
 # Copy the setup script and requirements file into the container

models/nllb.py

@@ -19,8 +19,15 @@ def nllb():
     #device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     # Load the tokenizer and model
+    # Set Hugging Face cache directory
     os.environ['HF_HOME'] = '/app/.cache/huggingface'
     os.environ['TRANSFORMERS_CACHE'] = '/app/.cache/huggingface'
+
+    # Create cache directory if it doesn't exist and set permissions
+    os.makedirs('/app/.cache/huggingface', exist_ok=True)
+    os.chmod('/app/.cache/huggingface', 0o777)
+
+    # Load models
     tokenizer = AutoTokenizer.from_pretrained("facebook/nllb-200-distilled-1.3B")
     model = AutoModelForSeq2SeqLM.from_pretrained("facebook/nllb-200-distilled-1.3B").to(device)
     # write done to the file named status.txt