adaptiveaiventures commited on
Commit
bc65857
·
verified ·
1 Parent(s): 7559f23

Update Dockerfile

Browse files
Files changed (1) hide show
  1. Dockerfile +4 -6
Dockerfile CHANGED
@@ -1,14 +1,12 @@
1
  FROM ghcr.io/huggingface/text-generation-inference:latest
2
 
3
- USER root # Run container as root
 
4
 
5
- # Define Model
6
- ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
7
-
8
- # Set cache directory to a writable location
9
  ENV HF_HOME="/tmp"
10
  ENV TRANSFORMERS_CACHE="/tmp"
11
  ENV HUGGINGFACE_HUB_CACHE="/tmp"
12
 
13
  # Run the TGI server
14
- CMD ["--model-id", "adaptiveaiventures/Llama-32-1B-finetuned-aaivtest5", "--port", "8080", "--num-shard", "1", "--dtype", "bfloat16", "--max-batch-prefill-tokens", "512", "--disable-custom-kernels"]
 
1
  FROM ghcr.io/huggingface/text-generation-inference:latest
2
 
3
+ # Define a pre-existing Hugging Face model
4
+ ENV MODEL_ID="gpt2"
5
 
6
+ # Use a writable cache directory
 
 
 
7
  ENV HF_HOME="/tmp"
8
  ENV TRANSFORMERS_CACHE="/tmp"
9
  ENV HUGGINGFACE_HUB_CACHE="/tmp"
10
 
11
  # Run the TGI server
12
+ CMD ["--model-id", "gpt2", "--port", "8080", "--num-shard", "1"]