RuntimeError: Expected to mark a variable ready only once.

Hi!

I tried to use peft model with Trainer, got the next error, I used gradient_checkpointing:

RuntimeError: Expected to mark a variable ready only once. This error is caused by one of the following reasons: 1) Use of a module parameter outside the `forward` function. Please make sure model parameters are not shared across multiple concurrent forward-backward passes. or try to use _set_static_graph() as a workaround if this module graph does not change during training loop.2) Reused parameters in multiple reentrant backward passes. For example, if you use multiple `checkpoint` functions to wrap the same part of your model, it would result in the same set of parameters been used by different reentrant backward passes multiple times, and hence marking a variable ready multiple times. DDP does not support such use cases in default. You can try to use _set_static_graph() as a workaround if your module graph does not change over iterations.

Code:

tokenizer_kwargs = {
      "cache_dir": None,
      "use_fast": False,
      "revision": 'main',
      "use_auth_token": None,
  }

tokenizer = AutoTokenizer.from_pretrained("facebook/opt-125m", **tokenizer_kwargs)

model = AutoModelForCausalLM.from_pretrained(
            "facebook/opt-125m",
            cache_dir=None,
            revision='main',
            use_auth_token=None,
            torch_dtype="auto"
        )

model.enable_input_require_grads()

config = LoraConfig(r=16, lora_alpha=32, target_modules=["q_proj", "v_proj"], lora_dropout=0.05, bias="none", 
task_type="CAUSAL_LM")

model = get_peft_model(model, config)

trainer = Trainer(
        model=model,
        args=training_args,
        train_dataset=train_dataset,
        eval_dataset=eval_dataset,
        tokenizer=tokenizer,
        data_collator=data_collator,
        compute_metrics=compute_metrics if training_args.do_eval and not is_torch_tpu_available() else None,
        preprocess_logits_for_metrics=preprocess_logits_for_metrics
        if training_args.do_eval and not is_torch_tpu_available()
        else None,
    )

training_args =(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.95,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=None,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=1000000000,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'fsdp_min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
gradient_accumulation_steps=1,
gradient_checkpointing=True,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=1e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=0,
log_level=passive,
log_level_replica=warning,
log_on_each_node=False,
logging_dir=output/runs/Apr12_13-04-09_bms-airtrunk-b-g1v2-app-10-169-60-152,
logging_first_step=True,
logging_nan_inf_filter=False,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=constant_with_warmup,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=output,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=16,
per_device_train_batch_size=32,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=['tensorboard'],
resume_from_checkpoint=None,
run_name=output,
save_on_each_node=False,
save_safetensors=False,
save_steps=100,
save_strategy=steps,
save_total_limit=3,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.1,
warmup_steps=0,
weight_decay=0.1,
xpu_backend=None,
)

huggingface / peft

RuntimeError: Expected to mark a variable ready only once. #301