Update Dockerfile

Dockerfile

@@ -3,8 +3,5 @@ FROM ghcr.io/huggingface/text-generation-inference:latest
 # Define the model to use
 ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
 
-# Set the number of GPU shards (adjust based on GPU availability)
-ENV NUM_SHARD=1
-
-# Run the TGI server
-CMD ["--model-id", "${MODEL_ID}", "--port", "8080", "--num-shard", "${NUM_SHARD}"]
+# Set the number of GPU shards (1 if using CPU, 2+ if using multiple GPUs)
+CMD ["--model-id", "adaptiveaiventures/Llama-2-7b-chat-finetune", "--port", "8080", "--num-shard", "1"]