End of training
Browse files- README.md +81 -0
- config.json +109 -0
- model.safetensors +3 -0
- preprocessor_config.json +10 -0
- trainer_state.json +348 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: facebook/wav2vec2-xls-r-300m
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: wav2vec-arabic-V2-50
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# wav2vec-arabic-V2-50
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3408
|
| 21 |
+
- Wer: 0.0460
|
| 22 |
+
- Per: 0.0347
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0001
|
| 42 |
+
- train_batch_size: 2
|
| 43 |
+
- eval_batch_size: 2
|
| 44 |
+
- seed: 42
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- num_epochs: 20
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Per |
|
| 53 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
|
| 54 |
+
| 10.1558 | 1.0 | 818 | 3.2108 | 1.0 | 1.0 |
|
| 55 |
+
| 3.0783 | 2.0 | 1636 | 2.0778 | 0.9260 | 0.9396 |
|
| 56 |
+
| 0.7602 | 3.0 | 2454 | 0.3836 | 0.1023 | 0.0870 |
|
| 57 |
+
| 0.2361 | 4.0 | 3272 | 0.3251 | 0.0668 | 0.0517 |
|
| 58 |
+
| 0.1602 | 5.0 | 4090 | 0.3240 | 0.0664 | 0.0515 |
|
| 59 |
+
| 0.1216 | 6.0 | 4908 | 0.3268 | 0.0673 | 0.0531 |
|
| 60 |
+
| 0.1078 | 7.0 | 5726 | 0.3501 | 0.0608 | 0.0465 |
|
| 61 |
+
| 0.0933 | 8.0 | 6544 | 0.3451 | 0.0538 | 0.0402 |
|
| 62 |
+
| 0.0713 | 9.0 | 7362 | 0.3658 | 0.0539 | 0.0407 |
|
| 63 |
+
| 0.0687 | 10.0 | 8180 | 0.3106 | 0.0519 | 0.0386 |
|
| 64 |
+
| 0.0561 | 11.0 | 8998 | 0.3322 | 0.0529 | 0.0396 |
|
| 65 |
+
| 0.0516 | 12.0 | 9816 | 0.3243 | 0.0484 | 0.0361 |
|
| 66 |
+
| 0.0392 | 13.0 | 10634 | 0.3412 | 0.0475 | 0.0354 |
|
| 67 |
+
| 0.037 | 14.0 | 11452 | 0.3370 | 0.0477 | 0.0359 |
|
| 68 |
+
| 0.0318 | 15.0 | 12270 | 0.3250 | 0.0466 | 0.0358 |
|
| 69 |
+
| 0.0291 | 16.0 | 13088 | 0.3451 | 0.0477 | 0.0359 |
|
| 70 |
+
| 0.025 | 17.0 | 13906 | 0.3713 | 0.0486 | 0.0368 |
|
| 71 |
+
| 0.0274 | 18.0 | 14724 | 0.3299 | 0.0459 | 0.0346 |
|
| 72 |
+
| 0.0208 | 19.0 | 15542 | 0.3451 | 0.0463 | 0.0349 |
|
| 73 |
+
| 0.0187 | 20.0 | 16360 | 0.3408 | 0.0460 | 0.0347 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.35.2
|
| 79 |
+
- Pytorch 2.1.0+cu121
|
| 80 |
+
- Datasets 2.16.1
|
| 81 |
+
- Tokenizers 0.15.0
|
config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.1,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 4096,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.1,
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_min_masks": 0,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_masks": 2,
|
| 69 |
+
"mask_time_prob": 0.075,
|
| 70 |
+
"model_type": "wav2vec2",
|
| 71 |
+
"num_adapter_layers": 3,
|
| 72 |
+
"num_attention_heads": 16,
|
| 73 |
+
"num_codevector_groups": 2,
|
| 74 |
+
"num_codevectors_per_group": 320,
|
| 75 |
+
"num_conv_pos_embedding_groups": 16,
|
| 76 |
+
"num_conv_pos_embeddings": 128,
|
| 77 |
+
"num_feat_extract_layers": 7,
|
| 78 |
+
"num_hidden_layers": 24,
|
| 79 |
+
"num_negatives": 100,
|
| 80 |
+
"output_hidden_size": 1024,
|
| 81 |
+
"pad_token_id": 37,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"tdnn_dilation": [
|
| 84 |
+
1,
|
| 85 |
+
2,
|
| 86 |
+
3,
|
| 87 |
+
1,
|
| 88 |
+
1
|
| 89 |
+
],
|
| 90 |
+
"tdnn_dim": [
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
512,
|
| 95 |
+
1500
|
| 96 |
+
],
|
| 97 |
+
"tdnn_kernel": [
|
| 98 |
+
5,
|
| 99 |
+
3,
|
| 100 |
+
3,
|
| 101 |
+
1,
|
| 102 |
+
1
|
| 103 |
+
],
|
| 104 |
+
"torch_dtype": "float32",
|
| 105 |
+
"transformers_version": "4.35.2",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 40,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11b952b04d7ce2d049c3db385d05039aa450f1af0adce162df0dcc31935cfe3e
|
| 3 |
+
size 1261971480
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,348 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3106290400028229,
|
| 3 |
+
"best_model_checkpoint": "nrshoudi/wav2vec-arabic-V2-50/checkpoint-8180",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16360,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"learning_rate": 5e-05,
|
| 14 |
+
"loss": 10.1558,
|
| 15 |
+
"step": 818
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"eval_loss": 3.2108113765716553,
|
| 20 |
+
"eval_per": 1.0,
|
| 21 |
+
"eval_runtime": 121.3546,
|
| 22 |
+
"eval_samples_per_second": 6.699,
|
| 23 |
+
"eval_steps_per_second": 3.354,
|
| 24 |
+
"eval_wer": 1.0,
|
| 25 |
+
"step": 818
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"learning_rate": 0.0001,
|
| 30 |
+
"loss": 3.0783,
|
| 31 |
+
"step": 1636
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 2.0,
|
| 35 |
+
"eval_loss": 2.077756881713867,
|
| 36 |
+
"eval_per": 0.9395552624140205,
|
| 37 |
+
"eval_runtime": 121.8972,
|
| 38 |
+
"eval_samples_per_second": 6.67,
|
| 39 |
+
"eval_steps_per_second": 3.339,
|
| 40 |
+
"eval_wer": 0.9260417152722692,
|
| 41 |
+
"step": 1636
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"learning_rate": 9.444444444444444e-05,
|
| 46 |
+
"loss": 0.7602,
|
| 47 |
+
"step": 2454
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"eval_loss": 0.38355985283851624,
|
| 52 |
+
"eval_per": 0.08696881183454254,
|
| 53 |
+
"eval_runtime": 123.229,
|
| 54 |
+
"eval_samples_per_second": 6.597,
|
| 55 |
+
"eval_steps_per_second": 3.303,
|
| 56 |
+
"eval_wer": 0.10232840278101815,
|
| 57 |
+
"step": 2454
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"learning_rate": 8.888888888888889e-05,
|
| 62 |
+
"loss": 0.2361,
|
| 63 |
+
"step": 3272
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.0,
|
| 67 |
+
"eval_loss": 0.325092613697052,
|
| 68 |
+
"eval_per": 0.05165834354094036,
|
| 69 |
+
"eval_runtime": 122.9982,
|
| 70 |
+
"eval_samples_per_second": 6.61,
|
| 71 |
+
"eval_steps_per_second": 3.309,
|
| 72 |
+
"eval_wer": 0.0667724324576548,
|
| 73 |
+
"step": 3272
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"learning_rate": 8.333333333333334e-05,
|
| 78 |
+
"loss": 0.1602,
|
| 79 |
+
"step": 4090
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 5.0,
|
| 83 |
+
"eval_loss": 0.3239767551422119,
|
| 84 |
+
"eval_per": 0.05151700744370112,
|
| 85 |
+
"eval_runtime": 122.92,
|
| 86 |
+
"eval_samples_per_second": 6.614,
|
| 87 |
+
"eval_steps_per_second": 3.311,
|
| 88 |
+
"eval_wer": 0.06639914143063787,
|
| 89 |
+
"step": 4090
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"learning_rate": 7.777777777777778e-05,
|
| 94 |
+
"loss": 0.1216,
|
| 95 |
+
"step": 4908
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 6.0,
|
| 99 |
+
"eval_loss": 0.32677653431892395,
|
| 100 |
+
"eval_per": 0.05311881654574578,
|
| 101 |
+
"eval_runtime": 123.0237,
|
| 102 |
+
"eval_samples_per_second": 6.608,
|
| 103 |
+
"eval_steps_per_second": 3.308,
|
| 104 |
+
"eval_wer": 0.06733236899818021,
|
| 105 |
+
"step": 4908
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"learning_rate": 7.222222222222222e-05,
|
| 110 |
+
"loss": 0.1078,
|
| 111 |
+
"step": 5726
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 7.0,
|
| 115 |
+
"eval_loss": 0.35006746649742126,
|
| 116 |
+
"eval_per": 0.04649957599170828,
|
| 117 |
+
"eval_runtime": 123.221,
|
| 118 |
+
"eval_samples_per_second": 6.598,
|
| 119 |
+
"eval_steps_per_second": 3.303,
|
| 120 |
+
"eval_wer": 0.06079977602538379,
|
| 121 |
+
"step": 5726
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"learning_rate": 6.666666666666667e-05,
|
| 126 |
+
"loss": 0.0933,
|
| 127 |
+
"step": 6544
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 8.0,
|
| 131 |
+
"eval_loss": 0.34512361884117126,
|
| 132 |
+
"eval_per": 0.04021011966456233,
|
| 133 |
+
"eval_runtime": 123.1692,
|
| 134 |
+
"eval_samples_per_second": 6.601,
|
| 135 |
+
"eval_steps_per_second": 3.304,
|
| 136 |
+
"eval_wer": 0.053847230647193316,
|
| 137 |
+
"step": 6544
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"learning_rate": 6.111111111111112e-05,
|
| 142 |
+
"loss": 0.0713,
|
| 143 |
+
"step": 7362
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 9.0,
|
| 147 |
+
"eval_loss": 0.36584600806236267,
|
| 148 |
+
"eval_per": 0.04072835202110619,
|
| 149 |
+
"eval_runtime": 123.1989,
|
| 150 |
+
"eval_samples_per_second": 6.599,
|
| 151 |
+
"eval_steps_per_second": 3.304,
|
| 152 |
+
"eval_wer": 0.05389389202557043,
|
| 153 |
+
"step": 7362
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 10.0,
|
| 157 |
+
"learning_rate": 5.555555555555556e-05,
|
| 158 |
+
"loss": 0.0687,
|
| 159 |
+
"step": 8180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.0,
|
| 163 |
+
"eval_loss": 0.3106290400028229,
|
| 164 |
+
"eval_per": 0.03856119853010459,
|
| 165 |
+
"eval_runtime": 123.4148,
|
| 166 |
+
"eval_samples_per_second": 6.588,
|
| 167 |
+
"eval_steps_per_second": 3.298,
|
| 168 |
+
"eval_wer": 0.05188745275535439,
|
| 169 |
+
"step": 8180
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 11.0,
|
| 173 |
+
"learning_rate": 5e-05,
|
| 174 |
+
"loss": 0.0561,
|
| 175 |
+
"step": 8998
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 11.0,
|
| 179 |
+
"eval_loss": 0.3322308361530304,
|
| 180 |
+
"eval_per": 0.03959766324319231,
|
| 181 |
+
"eval_runtime": 124.0615,
|
| 182 |
+
"eval_samples_per_second": 6.553,
|
| 183 |
+
"eval_steps_per_second": 3.281,
|
| 184 |
+
"eval_wer": 0.052914003079650974,
|
| 185 |
+
"step": 8998
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 190 |
+
"loss": 0.0516,
|
| 191 |
+
"step": 9816
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 12.0,
|
| 195 |
+
"eval_loss": 0.32428082823753357,
|
| 196 |
+
"eval_per": 0.036087816828417975,
|
| 197 |
+
"eval_runtime": 123.5702,
|
| 198 |
+
"eval_samples_per_second": 6.579,
|
| 199 |
+
"eval_steps_per_second": 3.294,
|
| 200 |
+
"eval_wer": 0.04843451075544772,
|
| 201 |
+
"step": 9816
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 13.0,
|
| 205 |
+
"learning_rate": 3.888888888888889e-05,
|
| 206 |
+
"loss": 0.0392,
|
| 207 |
+
"step": 10634
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 13.0,
|
| 211 |
+
"eval_loss": 0.3411938548088074,
|
| 212 |
+
"eval_per": 0.035404692358428344,
|
| 213 |
+
"eval_runtime": 123.9404,
|
| 214 |
+
"eval_samples_per_second": 6.56,
|
| 215 |
+
"eval_steps_per_second": 3.284,
|
| 216 |
+
"eval_wer": 0.04745462180952825,
|
| 217 |
+
"step": 10634
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 14.0,
|
| 221 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 222 |
+
"loss": 0.037,
|
| 223 |
+
"step": 11452
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 14.0,
|
| 227 |
+
"eval_loss": 0.3370288014411926,
|
| 228 |
+
"eval_per": 0.035899368698765664,
|
| 229 |
+
"eval_runtime": 123.6678,
|
| 230 |
+
"eval_samples_per_second": 6.574,
|
| 231 |
+
"eval_steps_per_second": 3.291,
|
| 232 |
+
"eval_wer": 0.04768792870141384,
|
| 233 |
+
"step": 11452
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.0,
|
| 237 |
+
"learning_rate": 2.777777777777778e-05,
|
| 238 |
+
"loss": 0.0318,
|
| 239 |
+
"step": 12270
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.0,
|
| 243 |
+
"eval_loss": 0.32496118545532227,
|
| 244 |
+
"eval_per": 0.03580514463393951,
|
| 245 |
+
"eval_runtime": 123.6467,
|
| 246 |
+
"eval_samples_per_second": 6.575,
|
| 247 |
+
"eval_steps_per_second": 3.292,
|
| 248 |
+
"eval_wer": 0.04656805562036303,
|
| 249 |
+
"step": 12270
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 254 |
+
"loss": 0.0291,
|
| 255 |
+
"step": 13088
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.0,
|
| 259 |
+
"eval_loss": 0.34510985016822815,
|
| 260 |
+
"eval_per": 0.035946480731178745,
|
| 261 |
+
"eval_runtime": 123.7608,
|
| 262 |
+
"eval_samples_per_second": 6.569,
|
| 263 |
+
"eval_steps_per_second": 3.289,
|
| 264 |
+
"eval_wer": 0.04773459007979096,
|
| 265 |
+
"step": 13088
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 17.0,
|
| 269 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 270 |
+
"loss": 0.025,
|
| 271 |
+
"step": 13906
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 17.0,
|
| 275 |
+
"eval_loss": 0.3712596893310547,
|
| 276 |
+
"eval_per": 0.03679449731461415,
|
| 277 |
+
"eval_runtime": 123.927,
|
| 278 |
+
"eval_samples_per_second": 6.56,
|
| 279 |
+
"eval_steps_per_second": 3.284,
|
| 280 |
+
"eval_wer": 0.04862115626895618,
|
| 281 |
+
"step": 13906
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 18.0,
|
| 285 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 286 |
+
"loss": 0.0274,
|
| 287 |
+
"step": 14724
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 18.0,
|
| 291 |
+
"eval_loss": 0.32992804050445557,
|
| 292 |
+
"eval_per": 0.03458023179119947,
|
| 293 |
+
"eval_runtime": 123.4837,
|
| 294 |
+
"eval_samples_per_second": 6.584,
|
| 295 |
+
"eval_steps_per_second": 3.296,
|
| 296 |
+
"eval_wer": 0.04591479632308339,
|
| 297 |
+
"step": 14724
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 19.0,
|
| 301 |
+
"learning_rate": 5.555555555555556e-06,
|
| 302 |
+
"loss": 0.0208,
|
| 303 |
+
"step": 15542
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 19.0,
|
| 307 |
+
"eval_loss": 0.34512093663215637,
|
| 308 |
+
"eval_per": 0.03488646000188448,
|
| 309 |
+
"eval_runtime": 122.6539,
|
| 310 |
+
"eval_samples_per_second": 6.628,
|
| 311 |
+
"eval_steps_per_second": 3.318,
|
| 312 |
+
"eval_wer": 0.04628808735010032,
|
| 313 |
+
"step": 15542
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"learning_rate": 0.0,
|
| 318 |
+
"loss": 0.0187,
|
| 319 |
+
"step": 16360
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 20.0,
|
| 323 |
+
"eval_loss": 0.3407500386238098,
|
| 324 |
+
"eval_per": 0.03474512390464524,
|
| 325 |
+
"eval_runtime": 122.6729,
|
| 326 |
+
"eval_samples_per_second": 6.627,
|
| 327 |
+
"eval_steps_per_second": 3.318,
|
| 328 |
+
"eval_wer": 0.0459614577014605,
|
| 329 |
+
"step": 16360
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.0,
|
| 333 |
+
"step": 16360,
|
| 334 |
+
"total_flos": 5.249198170404504e+18,
|
| 335 |
+
"train_loss": 0.7594956291916901,
|
| 336 |
+
"train_runtime": 11598.3794,
|
| 337 |
+
"train_samples_per_second": 2.821,
|
| 338 |
+
"train_steps_per_second": 1.411
|
| 339 |
+
}
|
| 340 |
+
],
|
| 341 |
+
"logging_steps": 500,
|
| 342 |
+
"max_steps": 16360,
|
| 343 |
+
"num_train_epochs": 20,
|
| 344 |
+
"save_steps": 500,
|
| 345 |
+
"total_flos": 5.249198170404504e+18,
|
| 346 |
+
"trial_name": null,
|
| 347 |
+
"trial_params": null
|
| 348 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c44c883fb894de12e9732d9d2505d76493b7d58378b9debc52aac831b48bd83
|
| 3 |
+
size 4600
|