wandb/run-20220130_175417-3n44neww/files/config.yaml

wandb_version: 1

_n_gpu:
  desc: null
  value: 1
_name_or_path:
  desc: null
  value: facebook/wav2vec2-xls-r-300m
_wandb:
  desc: null
  value:
    cli_version: 0.12.9
    framework: huggingface
    huggingface_version: 4.17.0.dev0
    is_jupyter_run: false
    is_kaggle_kernel: false
    m:
    - 1: train/global_step
      6:
      - 3
    - 1: train/loss
      5: 1
      6:
      - 1
    - 1: train/learning_rate
      5: 1
      6:
      - 1
    - 1: train/epoch
      5: 1
      6:
      - 1
    - 1: eval/loss
      5: 1
      6:
      - 1
    - 1: eval/wer
      5: 1
      6:
      - 1
    - 1: eval/runtime
      5: 1
      6:
      - 1
    - 1: eval/samples_per_second
      5: 1
      6:
      - 1
    - 1: eval/steps_per_second
      5: 1
      6:
      - 1
    python_version: 3.8.8
    start_time: 1643565257
    t:
      1:
      - 1
      - 5
      - 11
      2:
      - 1
      - 5
      - 11
      3:
      - 1
      - 7
      - 13
      4: 3.8.8
      5: 0.12.9
      6: 4.17.0.dev0
      8:
      - 5
activation_dropout:
  desc: null
  value: 0.05
adafactor:
  desc: null
  value: false
adam_beta1:
  desc: null
  value: 0.9
adam_beta2:
  desc: null
  value: 0.999
adam_epsilon:
  desc: null
  value: 1.0e-08
adapter_kernel_size:
  desc: null
  value: 3
adapter_stride:
  desc: null
  value: 2
add_adapter:
  desc: null
  value: false
add_cross_attention:
  desc: null
  value: false
apply_spec_augment:
  desc: null
  value: true
architectures:
  desc: null
  value:
  - Wav2Vec2ForPreTraining
attention_dropout:
  desc: null
  value: 0.0
bad_words_ids:
  desc: null
  value: null
bf16:
  desc: null
  value: false
bf16_full_eval:
  desc: null
  value: false
bos_token_id:
  desc: null
  value: 1
chunk_size_feed_forward:
  desc: null
  value: 0
classifier_proj_size:
  desc: null
  value: 256
codevector_dim:
  desc: null
  value: 768
contrastive_logits_temperature:
  desc: null
  value: 0.1
conv_bias:
  desc: null
  value: true
conv_dim:
  desc: null
  value:
  - 512
  - 512
  - 512
  - 512
  - 512
  - 512
  - 512
conv_kernel:
  desc: null
  value:
  - 10
  - 3
  - 3
  - 3
  - 3
  - 2
  - 2
conv_stride:
  desc: null
  value:
  - 5
  - 2
  - 2
  - 2
  - 2
  - 2
  - 2
cross_attention_hidden_size:
  desc: null
  value: null
ctc_loss_reduction:
  desc: null
  value: mean
ctc_zero_infinity:
  desc: null
  value: false
dataloader_drop_last:
  desc: null
  value: false
dataloader_num_workers:
  desc: null
  value: 0
dataloader_pin_memory:
  desc: null
  value: true
ddp_bucket_cap_mb:
  desc: null
  value: None
ddp_find_unused_parameters:
  desc: null
  value: None
debug:
  desc: null
  value: '[]'
decoder_start_token_id:
  desc: null
  value: null
deepspeed:
  desc: null
  value: None
disable_tqdm:
  desc: null
  value: false
diversity_loss_weight:
  desc: null
  value: 0.1
diversity_penalty:
  desc: null
  value: 0.0
do_eval:
  desc: null
  value: true
do_predict:
  desc: null
  value: false
do_sample:
  desc: null
  value: false
do_stable_layer_norm:
  desc: null
  value: true
do_train:
  desc: null
  value: true
early_stopping:
  desc: null
  value: false
encoder_no_repeat_ngram_size:
  desc: null
  value: 0
eos_token_id:
  desc: null
  value: 2
eval_accumulation_steps:
  desc: null
  value: None
eval_batch_size:
  desc: null
  value: 64
eval_steps:
  desc: null
  value: 5
evaluation_strategy:
  desc: null
  value: steps
feat_extract_activation:
  desc: null
  value: gelu
feat_extract_dropout:
  desc: null
  value: 0.0
feat_extract_norm:
  desc: null
  value: layer
feat_proj_dropout:
  desc: null
  value: 0.0
feat_quantizer_dropout:
  desc: null
  value: 0.0
final_dropout:
  desc: null
  value: 0.0
finetuning_task:
  desc: null
  value: null
forced_bos_token_id:
  desc: null
  value: null
forced_eos_token_id:
  desc: null
  value: null
fp16:
  desc: null
  value: true
fp16_backend:
  desc: null
  value: auto
fp16_full_eval:
  desc: null
  value: false
fp16_opt_level:
  desc: null
  value: O1
gradient_accumulation_steps:
  desc: null
  value: 1
gradient_checkpointing:
  desc: null
  value: true
greater_is_better:
  desc: null
  value: false
group_by_length:
  desc: null
  value: true
half_precision_backend:
  desc: null
  value: amp
hidden_act:
  desc: null
  value: gelu
hidden_dropout:
  desc: null
  value: 0.0
hidden_size:
  desc: null
  value: 1024
hub_model_id:
  desc: null
  value: None
hub_strategy:
  desc: null
  value: every_save
hub_token:
  desc: null
  value: <HUB_TOKEN>
id2label:
  desc: null
  value:
    '0': LABEL_0
    '1': LABEL_1
ignore_data_skip:
  desc: null
  value: false
initializer_range:
  desc: null
  value: 0.02
intermediate_size:
  desc: null
  value: 4096
is_decoder:
  desc: null
  value: false
is_encoder_decoder:
  desc: null
  value: false
label2id:
  desc: null
  value:
    LABEL_0: 0
    LABEL_1: 1
label_names:
  desc: null
  value: None
label_smoothing_factor:
  desc: null
  value: 0.0
layer_norm_eps:
  desc: null
  value: 1.0e-05
layerdrop:
  desc: null
  value: 0.0
learning_rate:
  desc: null
  value: 0.0002
length_column_name:
  desc: null
  value: input_length
length_penalty:
  desc: null
  value: 1.0
load_best_model_at_end:
  desc: null
  value: true
local_rank:
  desc: null
  value: -1
log_level:
  desc: null
  value: -1
log_level_replica:
  desc: null
  value: -1
log_on_each_node:
  desc: null
  value: true
logging_dir:
  desc: null
  value: ./runs/Jan30_17-53-36_job-1abccd0a-3293-4ffe-8274-9e8f841f653f
logging_first_step:
  desc: null
  value: false
logging_nan_inf_filter:
  desc: null
  value: true
logging_steps:
  desc: null
  value: 5
logging_strategy:
  desc: null
  value: steps
lr_scheduler_type:
  desc: null
  value: linear
mask_feature_length:
  desc: null
  value: 10
mask_feature_min_masks:
  desc: null
  value: 0
mask_feature_prob:
  desc: null
  value: 0.25
mask_time_length:
  desc: null
  value: 10
mask_time_min_masks:
  desc: null
  value: 2
mask_time_prob:
  desc: null
  value: 0.6
max_grad_norm:
  desc: null
  value: 1.0
max_length:
  desc: null
  value: 20
max_steps:
  desc: null
  value: -1
metric_for_best_model:
  desc: null
  value: loss
min_length:
  desc: null
  value: 0
model_type:
  desc: null
  value: wav2vec2
mp_parameters:
  desc: null
  value: ''
no_cuda:
  desc: null
  value: false
no_repeat_ngram_size:
  desc: null
  value: 0
num_adapter_layers:
  desc: null
  value: 3
num_attention_heads:
  desc: null
  value: 16
num_beam_groups:
  desc: null
  value: 1
num_beams:
  desc: null
  value: 1
num_codevector_groups:
  desc: null
  value: 2
num_codevectors_per_group:
  desc: null
  value: 320
num_conv_pos_embedding_groups:
  desc: null
  value: 16
num_conv_pos_embeddings:
  desc: null
  value: 128
num_feat_extract_layers:
  desc: null
  value: 7
num_hidden_layers:
  desc: null
  value: 24
num_negatives:
  desc: null
  value: 100
num_return_sequences:
  desc: null
  value: 1
num_train_epochs:
  desc: null
  value: 4.0
optim:
  desc: null
  value: adamw_hf
output_attentions:
  desc: null
  value: false
output_dir:
  desc: null
  value: ./
output_hidden_size:
  desc: null
  value: 1024
output_hidden_states:
  desc: null
  value: false
output_scores:
  desc: null
  value: false
overwrite_output_dir:
  desc: null
  value: true
pad_token_id:
  desc: null
  value: 218
past_index:
  desc: null
  value: -1
per_device_eval_batch_size:
  desc: null
  value: 64
per_device_train_batch_size:
  desc: null
  value: 64
per_gpu_eval_batch_size:
  desc: null
  value: None
per_gpu_train_batch_size:
  desc: null
  value: None
prediction_loss_only:
  desc: null
  value: false
prefix:
  desc: null
  value: null
problem_type:
  desc: null
  value: null
proj_codevector_dim:
  desc: null
  value: 768
pruned_heads:
  desc: null
  value: {}
push_to_hub:
  desc: null
  value: true
push_to_hub_model_id:
  desc: null
  value: None
push_to_hub_organization:
  desc: null
  value: None
push_to_hub_token:
  desc: null
  value: <PUSH_TO_HUB_TOKEN>
remove_invalid_values:
  desc: null
  value: false
remove_unused_columns:
  desc: null
  value: true
repetition_penalty:
  desc: null
  value: 1.0
report_to:
  desc: null
  value: '[''wandb'']'
resume_from_checkpoint:
  desc: null
  value: None
return_dict:
  desc: null
  value: true
return_dict_in_generate:
  desc: null
  value: false
run_name:
  desc: null
  value: xls-r-300m-fr
save_on_each_node:
  desc: null
  value: false
save_steps:
  desc: null
  value: 500
save_strategy:
  desc: null
  value: steps
save_total_limit:
  desc: null
  value: 20
seed:
  desc: null
  value: 42
sep_token_id:
  desc: null
  value: null
sharded_ddp:
  desc: null
  value: '[]'
skip_memory_metrics:
  desc: null
  value: true
task_specific_params:
  desc: null
  value: null
tdnn_dilation:
  desc: null
  value:
  - 1
  - 2
  - 3
  - 1
  - 1
tdnn_dim:
  desc: null
  value:
  - 512
  - 512
  - 512
  - 512
  - 1500
tdnn_kernel:
  desc: null
  value:
  - 5
  - 3
  - 3
  - 1
  - 1
temperature:
  desc: null
  value: 1.0
tf32:
  desc: null
  value: None
tie_encoder_decoder:
  desc: null
  value: false
tie_word_embeddings:
  desc: null
  value: true
tokenizer_class:
  desc: null
  value: null
top_k:
  desc: null
  value: 50
top_p:
  desc: null
  value: 1.0
torch_dtype:
  desc: null
  value: float32
torchscript:
  desc: null
  value: false
tpu_metrics_debug:
  desc: null
  value: false
tpu_num_cores:
  desc: null
  value: None
train_batch_size:
  desc: null
  value: 64
transformers_version:
  desc: null
  value: 4.17.0.dev0
use_bfloat16:
  desc: null
  value: false
use_legacy_prediction_loop:
  desc: null
  value: false
use_weighted_layer_sum:
  desc: null
  value: false
vocab_size:
  desc: null
  value: 219
warmup_ratio:
  desc: null
  value: 0.0
warmup_steps:
  desc: null
  value: 2500
weight_decay:
  desc: null
  value: 0.0
xpu_backend:
  desc: null
  value: None
xvector_output_dim:
  desc: null
  value: 512