Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-26 01:30:20 +01:00
Code Issues Packages Projects Releases Wiki Activity

clean up duplicated code

Browse Source
This commit is contained in:
Wojtek Kowaluk 2023-03-18 02:27:26 +01:00
parent 30939e2aee
commit 7994b580d5
1 changed files with 10 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
modules/models.py
View File

@ -46,15 +46,17 @@ def load_model(model_name):
if not any([shared.args.cpu, shared.args.load_in_8bit, shared.args.gptq_bits, shared.args.auto_devices, shared.args.disk, shared.args.gpu_memory is not None, shared.args.cpu_memory is not None, shared.args.deepspeed, shared.args.flexgen, shared.is_RWKV]): if not any([shared.args.cpu, shared.args.load_in_8bit, shared.args.gptq_bits, shared.args.auto_devices, shared.args.disk, shared.args.gpu_memory is not None, shared.args.cpu_memory is not None, shared.args.deepspeed, shared.args.flexgen, shared.is_RWKV]):
if any(size in shared.model_name.lower() for size in ('13b', '20b', '30b')): if any(size in shared.model_name.lower() for size in ('13b', '20b', '30b')):
model = AutoModelForCausalLM.from_pretrained(Path(f"models/{shared.model_name}"), device_map='auto', load_in_8bit=True) model = AutoModelForCausalLM.from_pretrained(Path(f"models/{shared.model_name}"), device_map='auto', load_in_8bit=True)
if torch.has_mps: else:
model = AutoModelForCausalLM.from_pretrained( model = AutoModelForCausalLM.from_pretrained(
Path(f"models/{shared.model_name}"),low_cpu_mem_usage=True, Path(f"models/{shared.model_name}"),
torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16 low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16
) )
if torch.has_mps:
device = torch.device('mps') device = torch.device('mps')
model = model.to(device) model = model.to(device)
else: else:
model = AutoModelForCausalLM.from_pretrained(Path(f"models/{shared.model_name}"), low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16).cuda() model = model.cuda()
# FlexGen # FlexGen
elif shared.args.flexgen: elif shared.args.flexgen: