__cached__setup_devices: !!python/object/apply:torch.device
- cuda
- 0
_n_gpu: 1
accelerator_config: !!python/object:transformers.trainer_pt_utils.AcceleratorConfig
  dispatch_batches: null
  even_batches: true
  gradient_accumulation_kwargs: null
  split_batches: false
  use_seedable_sampler: true
adafactor: false
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1.0e-08
auto_find_batch_size: false
bf16: false
bf16_full_eval: false
data_seed: null
dataloader_drop_last: false
dataloader_num_workers: 0
dataloader_persistent_workers: false
dataloader_pin_memory: true
dataloader_prefetch_factor: null
ddp_backend: null
ddp_broadcast_buffers: null
ddp_bucket_cap_mb: null
ddp_find_unused_parameters: null
ddp_timeout: 1800
debug: []
deepspeed: null
deepspeed_plugin: null
disable_tqdm: false
dispatch_batches: null
distributed_state: !!python/object:accelerate.state.PartialState
  _cpu: false
  backend: null
  debug: false
  device: !!python/object/apply:torch.device
  - cuda
  distributed_type: !!python/object/apply:accelerate.utils.dataclasses.DistributedType
  - 'NO'
  fork_launched: false
  local_process_index: 0
  num_processes: 1
  process_index: 0
do_eval: true
do_predict: false
do_train: false
eval_accumulation_steps: null
eval_delay: 0
eval_do_concat_batches: true
eval_steps: null
evaluation_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy
- epoch
fp16: true
fp16_backend: auto
fp16_full_eval: false
fp16_opt_level: O1
fsdp: []
fsdp_config:
  min_num_params: 0
  xla: false
  xla_fsdp_grad_ckpt: false
  xla_fsdp_v2: false
fsdp_min_num_params: 0
fsdp_transformer_layer_cls_to_wrap: null
full_determinism: false
generation_config: null
generation_max_length: null
generation_num_beams: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
gradient_checkpointing_kwargs: null
greater_is_better: false
group_by_length: false
half_precision_backend: auto
hub_always_push: false
hub_model_id: donut_experiment_bayesian_trial_20
hub_private_repo: false
hub_strategy: !!python/object/apply:transformers.trainer_utils.HubStrategy
- every_save
hub_token: null
ignore_data_skip: false
include_inputs_for_metrics: false
include_num_input_tokens_seen: false
include_tokens_per_second: false
jit_mode_eval: false
label_names: null
label_smoothing_factor: 0.0
learning_rate: 1.4151037707088747e-05
length_column_name: length
load_best_model_at_end: true
local_rank: 0
log_level: passive
log_level_replica: warning
log_on_each_node: true
logging_dir: model_runs/donut_experiment_bayesian_trial_20/runs/Jul01_17-11-13_ip-172-16-165-117.ec2.internal
logging_first_step: false
logging_nan_inf_filter: true
logging_steps: 100
logging_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy
- steps
lr_scheduler_kwargs: {}
lr_scheduler_type: !!python/object/apply:transformers.trainer_utils.SchedulerType
- linear
max_grad_norm: 1.0
max_steps: -1
metric_for_best_model: loss
mp_parameters: ''
neftune_noise_alpha: null
no_cuda: false
num_train_epochs: 4
optim: !!python/object/apply:transformers.training_args.OptimizerNames
- adamw_torch
optim_args: null
optim_target_modules: null
output_dir: model_runs/donut_experiment_bayesian_trial_20
overwrite_output_dir: false
past_index: -1
per_device_eval_batch_size: 1
per_device_train_batch_size: 1
per_gpu_eval_batch_size: null
per_gpu_train_batch_size: null
predict_with_generate: true
prediction_loss_only: false
push_to_hub: true
push_to_hub_model_id: null
push_to_hub_organization: null
push_to_hub_token: null
ray_scope: last
remove_unused_columns: true
report_to:
- tensorboard
resume_from_checkpoint: null
run_name: model_runs/donut_experiment_bayesian_trial_20
save_on_each_node: false
save_only_model: false
save_safetensors: true
save_steps: 500
save_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy
- epoch
save_total_limit: 2
seed: 42
skip_memory_metrics: true
sortish_sampler: false
split_batches: null
tf32: null
torch_compile: false
torch_compile_backend: null
torch_compile_mode: null
torchdynamo: null
tpu_metrics_debug: false
tpu_num_cores: null
use_cpu: false
use_ipex: false
use_legacy_prediction_loop: false
use_mps_device: false
warmup_ratio: 0.0
warmup_steps: 0
weight_decay: 0.00975620385252877