S^(h8 d dl Z d dlmZmZ d dlmZ ddlmZ ddlm Z m
Z
m Z m Z e j e Z e
r
d dlZddlmZ e G d d e Zy)
N) dataclassfield)Optional )TrainingArguments)cached_propertyis_tf_availableloggingrequires_backends)kerasc l e Zd ZU dZdZ edddi Zee e d< edddi Z
ee e d < eddd
i Z ee e d < ed dd
i Z e
e d< edddi Zee d< ededef fd Zedd Zedefd Zed Zedefd Zedefd Zedefd Zy) TFTrainingArgumentsa!
TrainingArguments is the subset of the arguments we use in our example scripts **which relate to the training loop
itself**.
Using [`HfArgumentParser`] we can turn this class into
[argparse](https://docs.python.org/3/library/argparse#module-argparse) arguments that can be specified on the
command line.
Parameters:
output_dir (`str`):
The output directory where the model predictions and checkpoints will be written.
overwrite_output_dir (`bool`, *optional*, defaults to `False`):
If `True`, overwrite the content of the output directory. Use this to continue training if `output_dir`
points to a checkpoint directory.
do_train (`bool`, *optional*, defaults to `False`):
Whether to run training or not. This argument is not directly used by [`Trainer`], it's intended to be used
by your training/evaluation scripts instead. See the [example
scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
do_eval (`bool`, *optional*):
Whether to run evaluation on the validation set or not. Will be set to `True` if `eval_strategy` is
different from `"no"`. This argument is not directly used by [`Trainer`], it's intended to be used by your
training/evaluation scripts instead. See the [example
scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
do_predict (`bool`, *optional*, defaults to `False`):
Whether to run predictions on the test set or not. This argument is not directly used by [`Trainer`], it's
intended to be used by your training/evaluation scripts instead. See the [example
scripts](https://github.com/huggingface/transformers/tree/main/examples) for more details.
eval_strategy (`str` or [`~trainer_utils.IntervalStrategy`], *optional*, defaults to `"no"`):
The evaluation strategy to adopt during training. Possible values are:
- `"no"`: No evaluation is done during training.
- `"steps"`: Evaluation is done (and logged) every `eval_steps`.
- `"epoch"`: Evaluation is done at the end of each epoch.
per_device_train_batch_size (`int`, *optional*, defaults to 8):
The batch size per GPU/TPU core/CPU for training.
per_device_eval_batch_size (`int`, *optional*, defaults to 8):
The batch size per GPU/TPU core/CPU for evaluation.
gradient_accumulation_steps (`int`, *optional*, defaults to 1):
Number of updates steps to accumulate the gradients for, before performing a backward/update pass.
When using gradient accumulation, one step is counted as one step with backward pass. Therefore, logging,
evaluation, save will be conducted every `gradient_accumulation_steps * xxx_step` training examples.
learning_rate (`float`, *optional*, defaults to 5e-5):
The initial learning rate for Adam.
weight_decay (`float`, *optional*, defaults to 0):
The weight decay to apply (if not zero).
adam_beta1 (`float`, *optional*, defaults to 0.9):
The beta1 hyperparameter for the Adam optimizer.
adam_beta2 (`float`, *optional*, defaults to 0.999):
The beta2 hyperparameter for the Adam optimizer.
adam_epsilon (`float`, *optional*, defaults to 1e-8):
The epsilon hyperparameter for the Adam optimizer.
max_grad_norm (`float`, *optional*, defaults to 1.0):
Maximum gradient norm (for gradient clipping).
num_train_epochs(`float`, *optional*, defaults to 3.0):
Total number of training epochs to perform.
max_steps (`int`, *optional*, defaults to -1):
If set to a positive number, the total number of training steps to perform. Overrides `num_train_epochs`.
For a finite dataset, training is reiterated through the dataset (if all data is exhausted) until
`max_steps` is reached.
warmup_ratio (`float`, *optional*, defaults to 0.0):
Ratio of total training steps used for a linear warmup from 0 to `learning_rate`.
warmup_steps (`int`, *optional*, defaults to 0):
Number of steps used for a linear warmup from 0 to `learning_rate`. Overrides any effect of `warmup_ratio`.
logging_dir (`str`, *optional*):
[TensorBoard](https://www.tensorflow.org/tensorboard) log directory. Will default to
*runs/**CURRENT_DATETIME_HOSTNAME***.
logging_strategy (`str` or [`~trainer_utils.IntervalStrategy`], *optional*, defaults to `"steps"`):
The logging strategy to adopt during training. Possible values are:
- `"no"`: No logging is done during training.
- `"epoch"`: Logging is done at the end of each epoch.
- `"steps"`: Logging is done every `logging_steps`.
logging_first_step (`bool`, *optional*, defaults to `False`):
Whether to log and evaluate the first `global_step` or not.
logging_steps (`int`, *optional*, defaults to 500):
Number of update steps between two logs if `logging_strategy="steps"`.
save_strategy (`str` or [`~trainer_utils.SaveStrategy`], *optional*, defaults to `"steps"`):
The checkpoint save strategy to adopt during training. Possible values are:
- `"no"`: No save is done during training.
- `"epoch"`: Save is done at the end of each epoch.
- `"steps"`: Save is done every `save_steps`.
save_steps (`int`, *optional*, defaults to 500):
Number of updates steps before two checkpoint saves if `save_strategy="steps"`.
save_total_limit (`int`, *optional*):
If a value is passed, will limit the total amount of checkpoints. Deletes the older checkpoints in
`output_dir`.
no_cuda (`bool`, *optional*, defaults to `False`):
Whether to not use CUDA even when it is available or not.
seed (`int`, *optional*, defaults to 42):
Random seed that will be set at the beginning of training.
fp16 (`bool`, *optional*, defaults to `False`):
Whether to use 16-bit (mixed) precision training (through NVIDIA Apex) instead of 32-bit training.
fp16_opt_level (`str`, *optional*, defaults to 'O1'):
For `fp16` training, Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']. See details on
the [Apex documentation](https://nvidia.github.io/apex/amp).
local_rank (`int`, *optional*, defaults to -1):
During distributed training, the rank of the process.
tpu_num_cores (`int`, *optional*):
When training on TPU, the number of TPU cores (automatically passed by launcher script).
debug (`bool`, *optional*, defaults to `False`):
Whether to activate the trace to record computation graphs and profiling information or not.
dataloader_drop_last (`bool`, *optional*, defaults to `False`):
Whether to drop the last incomplete batch (if the length of the dataset is not divisible by the batch size)
or not.
eval_steps (`int`, *optional*, defaults to 1000):
Number of update steps before two evaluations.
past_index (`int`, *optional*, defaults to -1):
Some models like [TransformerXL](../model_doc/transformerxl) or :doc*XLNet <../model_doc/xlnet>* can make
use of the past hidden states for their predictions. If this argument is set to a positive int, the
`Trainer` will use the corresponding output (usually index 2) as the past state and feed it to the model at
the next training step under the keyword argument `mems`.
tpu_name (`str`, *optional*):
The name of the TPU the process is running on.
tpu_zone (`str`, *optional*):
The zone of the TPU the process is running on. If not specified, we will attempt to automatically detect
from metadata.
gcp_project (`str`, *optional*):
Google Cloud Project name for the Cloud TPU-enabled project. If not specified, we will attempt to
automatically detect from metadata.
run_name (`str`, *optional*):
A descriptor for the run. Notably used for wandb, mlflow, comet and swanlab logging.
xla (`bool`, *optional*):
Whether to activate the XLA compilation or not.
tfNhelpz Name of TPU)defaultmetadatatpu_namez Zone of TPUtpu_zonez!Name of Cloud TPU-enabled project gcp_projectg ?z,Power for the Polynomial decay LR scheduler.
poly_powerFz.Whether to activate the XLA compilation or notxlareturntf.distribute.Strategyc t | dg t j d t j j d }| j r t j j d | j r"t j j d }|S | j rKt j j j | j | j | j" }n(t j j j }|r| j r t j j d
t j j) | t j* j, j/ | t j j1 | }|S t3 | d k( r"t j j d }|S t3 | d k( r"t j j d
}|S t3 | d kD r t j j5 }|S t% d # t$ $ r+ | j rt' d| j d d }Y bw xY w)Nr z Tensorflow: setting up strategyGPU
mixed_float16z/cpu:0)device)zoneprojectzCouldn't connect to TPU !mixed_bfloat16r r z/gpu:0zJCannot find the proper strategy, please check your environment properties.)r loggerinfor configlist_physical_devicesfp16r mixed_precisionset_global_policyno_cuda
distributeOneDeviceStrategyr cluster_resolverTPUClusterResolverr r
ValueError RuntimeError experimental_connect_to_clustertpu experimentalinitialize_tpu_system TPUStrategylenMirroredStrategy)selfgpusstrategyr1 s [/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/training_args_tf.py_setup_strategyz#TFTrainingArguments._setup_strategy s $' 56yy..u5 99 ! ! 3 3O D <<}}66h6GHF C
== --88KK
DMM4CSCS L C --88KKMC 99));;##99#> ==44S9 Ta ==::(:K Ta ==::(:K
TQ ==99; !!mnn1
==&)A$-- PQ'RSS C
s A?I
0I>=I>c 4 t | dg | j S )z=
The strategy used for distributed training.
r )r r; r7 s r: r9 z TFTrainingArguments.strategy s
$'### c H t | dg | j j S )Y
The number of replicas (CPUs, GPUs or TPU cores) used in this training.
r )r r; num_replicas_in_syncr= s r:
n_replicasz TFTrainingArguments.n_replicas s"
$'##888r> c y)zH
Whether or not the current process should produce log.
F r= s r:
should_logz TFTrainingArguments.should_log s
r> c | j rt j d | j xs | j }|| j z S )zz
The actual batch size for training (may differ from `per_gpu_train_batch_size` in distributed training).
zUsing deprecated `--per_gpu_train_batch_size` argument which will be removed in a future version. Using `--per_device_train_batch_size` is preferred.)per_gpu_train_batch_sizer" warningper_device_train_batch_sizerB r7 per_device_batch_sizes r: train_batch_sizez$TFTrainingArguments.train_batch_size sG
( ( NNO
!% = = aAaAa $t66r> c | j rt j d | j xs | j }|| j z S )z{
The actual batch size for evaluation (may differ from `per_gpu_eval_batch_size` in distributed training).
zUsing deprecated `--per_gpu_eval_batch_size` argument which will be removed in a future version. Using `--per_device_eval_batch_size` is preferred.)per_gpu_eval_batch_sizer" rH per_device_eval_batch_sizerB rJ s r: eval_batch_sizez#TFTrainingArguments.eval_batch_size sG
' ' NNN
!% < < _@_@_ $t66r> c | t | dg t j dt | j j
S )r@ r zaThe n_gpu argument is deprecated and will be removed in a future version, use n_replicas instead.)r warningswarn
FutureWarningr; rA r= s r: n_gpuzTFTrainingArguments.n_gpu! s6
$'
o
##888r> )r r )__name__
__module__ __qualname____doc__ frameworkr r r str__annotations__r r r floatr boolr tupleintr; propertyr9 rB rE rL rP rU rD r> r: r r sf EN I #-( Hhsm
$-( Hhsm
"'=>"K#
HIJ
ev7g.hiCi.'?'D!E . .` $ $ 9C 9 9
7#
7
7
7
7
7 9s 9 9r> r )rR dataclassesr r typingr
training_argsr utilsr r r
r
get_loggerrV r"
tensorflowr modeling_tf_utilsr r rD r> r: ri sX ( , O O
H %( K9+ K9 K9r>