text-generation-webui/modules/exllama_hf.py

107 lines
4.3 KiB
Python
Raw Normal View History

import os
from pathlib import Path
from typing import Any, Dict, Optional, Union
import torch
from torch.nn import CrossEntropyLoss
from transformers import GenerationConfig, PretrainedConfig, PreTrainedModel
from transformers.modeling_outputs import CausalLMOutputWithPast
from modules import shared
from modules.logging_colors import logger
try:
from exllama.model import ExLlama, ExLlamaCache, ExLlamaConfig
except:
logger.warning('Exllama module failed to load. Will attempt to load from repositories.')
try:
from modules.relative_imports import RelativeImport
with RelativeImport("repositories/exllama"):
from model import ExLlama, ExLlamaCache, ExLlamaConfig
except:
logger.error("Could not find repositories/exllama/. Make sure that exllama is cloned inside repositories/ and is up to date.")
raise
class ExllamaHF(PreTrainedModel):
def __init__(self, config: ExLlamaConfig):
super().__init__(PretrainedConfig())
self.ex_config = config
self.ex_model = ExLlama(self.ex_config)
self.generation_config = GenerationConfig()
2023-06-25 23:10:13 -04:00
self.lora = None
def _validate_model_class(self):
pass
def _validate_model_kwargs(self, model_kwargs: Dict[str, Any]):
pass
def prepare_inputs_for_generation(self, input_ids, **kwargs):
return {'input_ids': input_ids, **kwargs}
@property
def device(self) -> torch.device:
return torch.device(0)
def __call__(self, *args, **kwargs):
# TODO: Some decoding methods (such as Contrastive Search) may not work at this time
assert len(args) == 0, 'no *args should be passed to forward'
use_cache = kwargs.get('use_cache', True)
labels = kwargs.get('labels', None)
seq = kwargs['input_ids'][0].tolist()
cache = kwargs['past_key_values'] if 'past_key_values' in kwargs else None
if cache is None:
cache = ExLlamaCache(self.ex_model)
2023-06-25 23:10:13 -04:00
self.ex_model.forward(torch.tensor([seq[:-1]], dtype=torch.long), cache, preprocess_only=True, lora=self.lora)
2023-06-25 23:10:13 -04:00
logits = self.ex_model.forward(torch.tensor([seq[-1:]], dtype=torch.long), cache, lora=self.lora).to(kwargs['input_ids'].device)
loss = None
if labels is not None:
# Shift so that tokens < n predict n
shift_logits = logits[..., :-1, :].contiguous()
shift_labels = labels[..., 1:].contiguous()
# Flatten the tokens
loss_fct = CrossEntropyLoss()
shift_logits = shift_logits.view(-1, logits.shape[-1])
shift_labels = shift_labels.view(-1)
# Enable model parallelism
shift_labels = shift_labels.to(shift_logits.device)
loss = loss_fct(shift_logits, shift_labels)
return CausalLMOutputWithPast(logits=logits, past_key_values=cache if use_cache else None)
@classmethod
def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union[str, os.PathLike]], *model_args, **kwargs):
assert len(model_args) == 0 and len(kwargs) == 0, "extra args is currently not supported"
if isinstance(pretrained_model_name_or_path, str):
pretrained_model_name_or_path = Path(pretrained_model_name_or_path)
pretrained_model_name_or_path = Path(f'{shared.args.model_dir}') / Path(pretrained_model_name_or_path)
config = ExLlamaConfig(pretrained_model_name_or_path / 'config.json')
# from 'oobabooga/text-generation-webui/modules/exllama.py'
weight_path = None
for ext in ['.safetensors', '.pt', '.bin']:
found = list(pretrained_model_name_or_path.glob(f"*{ext}"))
if len(found) > 0:
weight_path = found[-1]
break
assert weight_path is not None, f'could not find weight in "{pretrained_model_name_or_path}"'
config.model_path = str(weight_path)
2023-06-25 21:49:26 -04:00
config.max_seq_len = shared.args.max_seq_len
config.compress_pos_emb = shared.args.compress_pos_emb
if shared.args.gpu_split:
config.set_auto_map(shared.args.gpu_split)
config.gpu_peer_fix = True
# This slowes down a bit but align better with autogptq generation.
# TODO: Should give user choice to tune the exllama config
# config.fused_attn = False
# config.fused_mlp_thd = 0
return ExllamaHF(config)