mmarch/text-generation-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Add "Save current settings for this model" button

Browse source
This commit is contained in:
oobabooga 2023-04-15 12:54:02 -03:00
parent b9dcba7762
commit ac189011cb
3 changed files with 40 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

15
modules/models.py
View file

@ -45,17 +45,14 @@ def load_model(model_name):
shared.is_RWKV = 'rwkv-' in model_name.lower()
shared.is_llamacpp = len(list(Path(f'{shared.args.model_dir}/{model_name}').glob('ggml*.bin'))) > 0
# Default settings
# Load the model in simple 16-bit mode by default
if not any([shared.args.cpu, shared.args.load_in_8bit, shared.args.wbits, 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, shared.is_llamacpp]):
if any(size in shared.model_name.lower() for size in ('13b', '20b', '30b')):
model = AutoModelForCausalLM.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}"), device_map='auto', load_in_8bit=True)
model = AutoModelForCausalLM.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}"), low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16)
if torch.has_mps:
device = torch.device('mps')
model = model.to(device)
else:
model = AutoModelForCausalLM.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}"), low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16)
if torch.has_mps:
device = torch.device('mps')
model = model.to(device)
else:
model = model.cuda()
model = model.cuda()
# FlexGen
elif shared.args.flexgen: