Matthias Reso 207d2f80e9 Make code-llama and hf-tgi inference runnable as module		1 anno fa
..
code_llama	207d2f80e9 Make code-llama and hf-tgi inference runnable as module	1 anno fa
hf_text_generation_inference	207d2f80e9 Make code-llama and hf-tgi inference runnable as module	1 anno fa
README.md	4c9cc7d223 Move modules into separate src folder	1 anno fa
__init__.py	207d2f80e9 Make code-llama and hf-tgi inference runnable as module	1 anno fa
__main__.py	207d2f80e9 Make code-llama and hf-tgi inference runnable as module	1 anno fa
chat_completion.py	5ac5d992ac Update imports in chat_completion	1 anno fa
chat_utils.py	cf678b9bf0 Adjust imports to package structure + cleaned up imports	1 anno fa
chats.json	4c9cc7d223 Move modules into separate src folder	1 anno fa
checkpoint_converter_fsdp_hf.py	cf678b9bf0 Adjust imports to package structure + cleaned up imports	1 anno fa
inference.py	cf678b9bf0 Adjust imports to package structure + cleaned up imports	1 anno fa
model_utils.py	4c9cc7d223 Move modules into separate src folder	1 anno fa
safety_utils.py	cf678b9bf0 Adjust imports to package structure + cleaned up imports	1 anno fa
samsum_prompt.txt	4c9cc7d223 Move modules into separate src folder	1 anno fa
vLLM_inference.py	cf678b9bf0 Adjust imports to package structure + cleaned up imports	1 anno fa

Inference

This folder contains inference examples for Llama 2. So far, we have provided support for three methods of inference:

inference script script provides support for Hugging Face accelerate, PEFT and FSDP fine tuned models.
vLLM_inference.py script takes advantage of vLLM's paged attention concept for low latency.
The hf-text-generation-inference folder contains information on Hugging Face Text Generation Inference (TGI).

For more in depth information on inference including inference safety checks and examples, see the inference documentation here.