huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-10 11:54:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Directly load GPTBigCode to specified device

Browse Source 
This PR directly load GPTBigCode to specified device, avoiding moving model between devices.
This commit is contained in:
Yang, Bo 2023-07-15 00:32:46 -07:00 committed by GitHub
parent c58a0c185b
commit 3e5165c3ed
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
server/text_generation_server/models/santacoder.py
View File

@ -51,14 +51,14 @@ class SantaCoder(CausalLM):
"pad_token": EOD,
}
)
model = AutoModelForCausalLM.from_pretrained(
model_id,
revision=revision,
torch_dtype=dtype,
load_in_8bit=quantize == "bitsandbytes",
trust_remote_code=trust_remote_code,
).to(device)
with device:
model = AutoModelForCausalLM.from_pretrained(
model_id,
revision=revision,
torch_dtype=dtype,
load_in_8bit=quantize == "bitsandbytes",
trust_remote_code=trust_remote_code,
)
super(CausalLM, self).__init__(
model=model,