Add model settings to the Models tab

This commit is contained in:
oobabooga 2023-04-12 17:09:56 -03:00
parent 4f7e88c043
commit 1566d8e344
3 changed files with 136 additions and 60 deletions

View file

@ -41,6 +41,7 @@ settings = {
'truncation_length': 2048,
'truncation_length_min': 0,
'truncation_length_max': 4096,
'mode': 'cai-chat',
'chat_prompt_size': 2048,
'chat_prompt_size_min': 0,
'chat_prompt_size_max': 2048,
@ -115,9 +116,6 @@ parser.add_argument('--wbits', type=int, default=0, help='GPTQ: Load a pre-quant
parser.add_argument('--model_type', type=str, help='GPTQ: Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported.')
parser.add_argument('--groupsize', type=int, default=-1, help='GPTQ: Group size.')
parser.add_argument('--pre_layer', type=int, default=0, help='GPTQ: The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models.')
parser.add_argument('--gptq-bits', type=int, default=0, help='DEPRECATED: use --wbits instead.')
parser.add_argument('--gptq-model-type', type=str, help='DEPRECATED: use --model_type instead.')
parser.add_argument('--gptq-pre-layer', type=int, default=0, help='DEPRECATED: use --pre_layer instead.')
# FlexGen
parser.add_argument('--flexgen', action='store_true', help='Enable the use of FlexGen offloading.')
@ -144,7 +142,7 @@ parser.add_argument("--gradio-auth-path", type=str, help='Set the gradio authent
args = parser.parse_args()
# Deprecation warnings for parameters that have been renamed
deprecated_dict = {'gptq_bits': ['wbits', 0], 'gptq_model_type': ['model_type', None], 'gptq_pre_layer': ['prelayer', 0]}
deprecated_dict = {}
for k in deprecated_dict:
if eval(f"args.{k}") != deprecated_dict[k][1]:
print(f"Warning: --{k} is deprecated and will be removed. Use --{deprecated_dict[k][0]} instead.")