Merge

Add support for HPU quantization and torch_compile (#207) #97

Job	Run time
Lint / lint	32s
Tests / bare-test (3.9)	15m 51s
Tests / getting-started-test (3.9)	3m 20s
Tests / bare-test (3.10)	15m 39s
Tests / docker-test (3.9)	19m 6s
Tests / bare-test (3.11)	15m 41s
Tests / docker-test (3.10)	17m 49s
Tests / getting-started-test (3.10)	2m 46s
Tests / docker-test (3.11)	17m 31s
Tests / setup-test (3.9)	2m 7s
Tests / getting-started-test (3.11)	2m 58s
Tests / setup-test (3.10)	2m 35s
Tests / setup-test (3.11)	2m 36s
Finetune / finetune (EleutherAI/gpt-j-6b)	28s
Finetune / finetune (meta-llama/Llama-2-7b-chat-hf)	22s
Inference / inference (gpt-j-6b)	25s
Finetune / finetune (mistralai/Mistral-7B-v0.1)	16s
Inference_Gaudi / inference (llama-2-7b-chat-hf)	8s
Finetune / finetune (google/gemma-2b)	18s
Inference / inference (mistral-7b-v0.1)	1s
Inference / inference (mpt-7b-ipex-llm)	1s
Inference / inference (llama-2-7b-chat-hf-vllm)	26s
Inference / inference (gemma-2b)	22s
Inference / inference (mpt-7b, nathan0/mpt-7b-deltatuner-model)	1s
	2h 1m 19s

Provide feedback