Open cococoda opened 4 months ago
05/30 18:04:29 - mmengine - WARNING - WARNING: command error: '[Errno 2] No such file or directory: '/root/autodl-tmp/ft/data/cutlass/CHANGELOG.md''! 05/30 18:04:29 - mmengine - WARNING - Arguments received: ['xtuner', 'train', 'internlm2_chat_7b_qlora_alpaca_e3_copy.py']. xtuner commands use the following syntax:
xtuner MODE MODE_ARGS ARGS
Where MODE (required) is one of ('list-cfg', 'copy-cfg', 'log-dataset', 'check-custom-dataset', 'train', 'test', 'chat', 'convert', 'preprocess', 'mmbench', 'eval_refcoco')
MODE_ARG (optional) is the argument for specific mode
ARGS (optional) are the arguments for specific command
Some usages for xtuner commands: (See more by using -h for specific command!)
same
运行xtuner train /root/autodl-tmp/ft/config/internlm2_chat_7b_qlora_alpaca_e3_copy.py --work-dir /root/autodl-tmp/ft/train时 `[2024-05-30 17:18:47,089] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) 05/30 17:18:49 - mmengine - WARNING - WARNING: command error: 'cannot import name 'packaging' from 'pkg_resources' (/root/miniconda3/envs/xtuner/lib/python3.10/site-packages/pkg_resources/init.py)'! 05/30 17:18:49 - mmengine - WARNING - Arguments received: ['xtuner', 'train', '/root/autodl-tmp/ft/config/internlm2_chat_7b_qlora_alpaca_e3_copy.py', '--work-dir', '/root/autodl-tmp/ft/train']. xtuner commands use the following syntax:
config文件 `# Copyright (c) OpenMMLab. All rights reserved. import torch from datasets import load_dataset from mmengine.dataset import DefaultSampler from mmengine.hooks import (CheckpointHook, DistSamplerSeedHook, IterTimerHook, LoggerHook, ParamSchedulerHook) from mmengine.optim import AmpOptimWrapper, CosineAnnealingLR, LinearLR from peft import LoraConfig from torch.optim import AdamW from transformers import (AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig)
from xtuner.dataset import process_hf_dataset from xtuner.dataset.collate_fns import default_collate_fn from xtuner.dataset.map_fns import openai_map_fn, template_map_fn_factory from xtuner.engine.hooks import (DatasetInfoHook, EvaluateChatHook, VarlenAttnArgsToMessageHubHook) from xtuner.engine.runner import TrainLoop from xtuner.model import SupervisedFinetune from xtuner.parallel.sequence import SequenceParallelSampler from xtuner.utils import PROMPT_TEMPLATE, SYSTEM_TEMPLATE
#######################################################################
PART 1 Settings
#######################################################################
Model
pretrained_model_name_or_path = '/root/autodl-tmp/RAG-langchain/models/internlm2-chat-7b' use_varlen_attn = False
Data
alpaca_en_path = '/root/autodl-tmp/ft/data/train_fold_1.json' prompt_template = PROMPT_TEMPLATE.internlm2_chat max_length = 1024 pack_to_max_length = True
parallel
sequence_parallel_size = 1
Scheduler & Optimizer
batch_size = 2 # per_device accumulative_counts = 16 accumulative_counts *= sequence_parallel_size dataloader_num_workers = 0 max_epochs = 3 optim_type = AdamW lr = 2e-4 betas = (0.9, 0.999) weight_decay = 0 max_norm = 1 # grad clip warmup_ratio = 0.03
Save
save_steps = 500 save_total_limit = 3 # Maximum checkpoints to keep (-1 means unlimited)
Evaluate the generation performance during the training
evaluation_freq = 300 SYSTEM = SYSTEM_TEMPLATE.alpaca evaluation_inputs = [ '请对以下新闻进行情绪分析,积极为1,消极为0', '判断该新闻情绪为正面还是负面,正面为1,负面为0' ]
#######################################################################
PART 2 Model & Tokenizer
####################################################################### tokenizer = dict( type=AutoTokenizer.from_pretrained, pretrained_model_name_or_path=pretrained_model_name_or_path, trust_remote_code=True, padding_side='right')
model = dict( type=SupervisedFinetune, use_varlen_attn=use_varlen_attn, llm=dict( type=AutoModelForCausalLM.from_pretrained, pretrained_model_name_or_path=pretrained_model_name_or_path, trust_remote_code=True, torch_dtype=torch.float16, quantization_config=dict( type=BitsAndBytesConfig, load_in_4bit=True, load_in_8bit=False, llm_int8_threshold=6.0, llm_int8_has_fp16_weight=False, bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, bnb_4bit_quant_type='nf4')), lora=dict( type=LoraConfig, r=64, lora_alpha=16, lora_dropout=0.1, bias='none', task_type='CAUSAL_LM'))
#######################################################################
PART 3 Dataset & Dataloader
####################################################################### alpaca_en = dict( type=process_hf_dataset, dataset=dict(type=load_dataset, path='json', data_files=dict(train=alpaca_en_path)), tokenizer=tokenizer, max_length=max_length, dataset_map_fn=openai_map_fn, template_map_fn=dict( type=template_map_fn_factory, template=prompt_template), remove_unused_columns=True, shuffle_before_pack=True, pack_to_max_length=pack_to_max_length, use_varlen_attn=use_varlen_attn)
sampler = SequenceParallelSampler \ if sequence_parallel_size > 1 else DefaultSampler train_dataloader = dict( batch_size=batch_size, num_workers=dataloader_num_workers, dataset=alpaca_en, sampler=dict(type=sampler, shuffle=True), collate_fn=dict(type=default_collate_fn, use_varlen_attn=use_varlen_attn))
#######################################################################
PART 4 Scheduler & Optimizer
#######################################################################
optimizer
optim_wrapper = dict( type=AmpOptimWrapper, optimizer=dict( type=optim_type, lr=lr, betas=betas, weight_decay=weight_decay), clip_grad=dict(max_norm=max_norm, error_if_nonfinite=False), accumulative_counts=accumulative_counts, loss_scale='dynamic', dtype='float16')
learning policy
More information: https://github.com/open-mmlab/mmengine/blob/main/docs/en/tutorials/param_scheduler.md # noqa: E501
param_scheduler = [ dict( type=LinearLR, start_factor=1e-5, by_epoch=True, begin=0, end=warmup_ratio max_epochs, convert_to_iter_based=True), dict( type=CosineAnnealingLR, eta_min=0.0, by_epoch=True, begin=warmup_ratio max_epochs, end=max_epochs, convert_to_iter_based=True) ]
train, val, test setting
train_cfg = dict(type=TrainLoop, max_epochs=max_epochs)
#######################################################################
PART 5 Runtime
#######################################################################
Log the dialogue periodically during the training process, optional
custom_hooks = [ dict(type=DatasetInfoHook, tokenizer=tokenizer), dict( type=EvaluateChatHook, tokenizer=tokenizer, every_n_iters=evaluation_freq, evaluation_inputs=evaluation_inputs, system=SYSTEM, prompt_template=prompt_template) ]
if use_varlen_attn: custom_hooks += [dict(type=VarlenAttnArgsToMessageHubHook)]
configure default hooks
default_hooks = dict(
record the time of every iteration.
)
configure environment
env_cfg = dict(
whether to enable cudnn benchmark
)
set visualizer
visualizer = None
set log level
log_level = 'INFO'
load from which checkpoint
load_from = None
whether to resume training from the loaded checkpoint
resume = False
Defaults to use random seed and disable
deterministic
randomness = dict(seed=None, deterministic=False)
set log processor
log_processor = dict(by_epoch=False) `