End of training
Browse files- README.md +82 -0
- all_results.json +9 -0
- config.json +116 -0
- model.safetensors +3 -0
- preprocessor_config.json +10 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +48 -0
- train_results.json +9 -0
- trainer_state.json +328 -0
- training_args.bin +3 -0
- vocab.json +42 -0
README.md
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: masoudmzb/wav2vec2-xlsr-multilingual-53-fa
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-large-xlsr-persian-asr-shemo_me7494
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# wav2vec2-large-xlsr-persian-asr-shemo_me7494
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [masoudmzb/wav2vec2-xlsr-multilingual-53-fa](https://huggingface.co/masoudmzb/wav2vec2-xlsr-multilingual-53-fa) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.6728
|
| 20 |
+
- Wer: 0.3286
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 1e-05
|
| 40 |
+
- train_batch_size: 8
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 500
|
| 48 |
+
- training_steps: 2000
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 1.8553 | 0.62 | 100 | 1.4126 | 0.4866 |
|
| 56 |
+
| 1.4083 | 1.25 | 200 | 1.0428 | 0.4366 |
|
| 57 |
+
| 1.1718 | 1.88 | 300 | 0.8683 | 0.4127 |
|
| 58 |
+
| 0.9919 | 2.5 | 400 | 0.7921 | 0.3919 |
|
| 59 |
+
| 0.9493 | 3.12 | 500 | 0.7676 | 0.3744 |
|
| 60 |
+
| 0.9414 | 3.75 | 600 | 0.7247 | 0.3695 |
|
| 61 |
+
| 0.8897 | 4.38 | 700 | 0.7202 | 0.3598 |
|
| 62 |
+
| 0.8716 | 5.0 | 800 | 0.7096 | 0.3546 |
|
| 63 |
+
| 0.8467 | 5.62 | 900 | 0.7023 | 0.3499 |
|
| 64 |
+
| 0.8227 | 6.25 | 1000 | 0.6994 | 0.3411 |
|
| 65 |
+
| 0.855 | 6.88 | 1100 | 0.6883 | 0.3432 |
|
| 66 |
+
| 0.8457 | 7.5 | 1200 | 0.6773 | 0.3426 |
|
| 67 |
+
| 0.7614 | 8.12 | 1300 | 0.6913 | 0.3344 |
|
| 68 |
+
| 0.8127 | 8.75 | 1400 | 0.6827 | 0.3335 |
|
| 69 |
+
| 0.8443 | 9.38 | 1500 | 0.6725 | 0.3356 |
|
| 70 |
+
| 0.7548 | 10.0 | 1600 | 0.6759 | 0.3318 |
|
| 71 |
+
| 0.7839 | 10.62 | 1700 | 0.6773 | 0.3286 |
|
| 72 |
+
| 0.7912 | 11.25 | 1800 | 0.6748 | 0.3286 |
|
| 73 |
+
| 0.8238 | 11.88 | 1900 | 0.6735 | 0.3297 |
|
| 74 |
+
| 0.7618 | 12.5 | 2000 | 0.6728 | 0.3286 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
### Framework versions
|
| 78 |
+
|
| 79 |
+
- Transformers 4.35.2
|
| 80 |
+
- Pytorch 2.1.0+cu118
|
| 81 |
+
- Datasets 2.15.0
|
| 82 |
+
- Tokenizers 0.15.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.5,
|
| 3 |
+
"total_flos": 9.605345431463184e+18,
|
| 4 |
+
"train_loss": 0.9391711463928223,
|
| 5 |
+
"train_runtime": 9697.5717,
|
| 6 |
+
"train_samples": 2554,
|
| 7 |
+
"train_samples_per_second": 3.3,
|
| 8 |
+
"train_steps_per_second": 0.206
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "masoudmzb/wav2vec2-xlsr-multilingual-53-fa",
|
| 3 |
+
"activation_dropout": 0.09216,
|
| 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": 0.05316,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 256,
|
| 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": true,
|
| 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": 0.01249,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.01941,
|
| 58 |
+
"hidden_size": 1024,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 4096,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.01377,
|
| 63 |
+
"mask_channel_length": 10,
|
| 64 |
+
"mask_channel_min_space": 1,
|
| 65 |
+
"mask_channel_other": 0.0,
|
| 66 |
+
"mask_channel_prob": 0.0,
|
| 67 |
+
"mask_channel_selection": "static",
|
| 68 |
+
"mask_feature_length": 10,
|
| 69 |
+
"mask_feature_min_masks": 0,
|
| 70 |
+
"mask_feature_prob": 0.0,
|
| 71 |
+
"mask_time_length": 10,
|
| 72 |
+
"mask_time_min_masks": 2,
|
| 73 |
+
"mask_time_min_space": 1,
|
| 74 |
+
"mask_time_other": 0.0,
|
| 75 |
+
"mask_time_prob": 0.04529,
|
| 76 |
+
"mask_time_selection": "static",
|
| 77 |
+
"model_type": "wav2vec2",
|
| 78 |
+
"num_adapter_layers": 3,
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_codevector_groups": 2,
|
| 81 |
+
"num_codevectors_per_group": 320,
|
| 82 |
+
"num_conv_pos_embedding_groups": 16,
|
| 83 |
+
"num_conv_pos_embeddings": 128,
|
| 84 |
+
"num_feat_extract_layers": 7,
|
| 85 |
+
"num_hidden_layers": 24,
|
| 86 |
+
"num_negatives": 100,
|
| 87 |
+
"output_hidden_size": 1024,
|
| 88 |
+
"pad_token_id": 0,
|
| 89 |
+
"proj_codevector_dim": 256,
|
| 90 |
+
"tdnn_dilation": [
|
| 91 |
+
1,
|
| 92 |
+
2,
|
| 93 |
+
3,
|
| 94 |
+
1,
|
| 95 |
+
1
|
| 96 |
+
],
|
| 97 |
+
"tdnn_dim": [
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
1500
|
| 103 |
+
],
|
| 104 |
+
"tdnn_kernel": [
|
| 105 |
+
5,
|
| 106 |
+
3,
|
| 107 |
+
3,
|
| 108 |
+
1,
|
| 109 |
+
1
|
| 110 |
+
],
|
| 111 |
+
"torch_dtype": "float32",
|
| 112 |
+
"transformers_version": "4.35.2",
|
| 113 |
+
"use_weighted_layer_sum": false,
|
| 114 |
+
"vocab_size": 40,
|
| 115 |
+
"xvector_output_dim": 512
|
| 116 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:131312a00c902f3cc61f5a4f9a853a7658f9a9a29d8db265ec4cb717bb47c027
|
| 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": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<pad>",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": true,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": false
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": true,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": false
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": true,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": true,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": false
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": true,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": true,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": false
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"bos_token": "<s>",
|
| 37 |
+
"clean_up_tokenization_spaces": true,
|
| 38 |
+
"do_lower_case": false,
|
| 39 |
+
"eos_token": "</s>",
|
| 40 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 41 |
+
"pad_token": "<pad>",
|
| 42 |
+
"processor_class": "Wav2Vec2Processor",
|
| 43 |
+
"replace_word_delimiter_char": " ",
|
| 44 |
+
"target_lang": null,
|
| 45 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 46 |
+
"unk_token": "<unk>",
|
| 47 |
+
"word_delimiter_token": "|"
|
| 48 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.5,
|
| 3 |
+
"total_flos": 9.605345431463184e+18,
|
| 4 |
+
"train_loss": 0.9391711463928223,
|
| 5 |
+
"train_runtime": 9697.5717,
|
| 6 |
+
"train_samples": 2554,
|
| 7 |
+
"train_samples_per_second": 3.3,
|
| 8 |
+
"train_steps_per_second": 0.206
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,328 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3285630841121495,
|
| 3 |
+
"best_model_checkpoint": "/content/drive/MyDrive/wav2vec2-large-xlsr-persian-asr-shemo_me7494/checkpoint-1700",
|
| 4 |
+
"epoch": 12.5,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.62,
|
| 13 |
+
"learning_rate": 1.98e-06,
|
| 14 |
+
"loss": 1.8553,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.62,
|
| 19 |
+
"eval_loss": 1.4126451015472412,
|
| 20 |
+
"eval_runtime": 35.7389,
|
| 21 |
+
"eval_samples_per_second": 7.947,
|
| 22 |
+
"eval_steps_per_second": 1.007,
|
| 23 |
+
"eval_wer": 0.48656542056074764,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.25,
|
| 28 |
+
"learning_rate": 3.980000000000001e-06,
|
| 29 |
+
"loss": 1.4083,
|
| 30 |
+
"step": 200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.25,
|
| 34 |
+
"eval_loss": 1.0428400039672852,
|
| 35 |
+
"eval_runtime": 35.3623,
|
| 36 |
+
"eval_samples_per_second": 8.031,
|
| 37 |
+
"eval_steps_per_second": 1.018,
|
| 38 |
+
"eval_wer": 0.4366238317757009,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.88,
|
| 43 |
+
"learning_rate": 5.98e-06,
|
| 44 |
+
"loss": 1.1718,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.88,
|
| 49 |
+
"eval_loss": 0.86826491355896,
|
| 50 |
+
"eval_runtime": 34.9958,
|
| 51 |
+
"eval_samples_per_second": 8.115,
|
| 52 |
+
"eval_steps_per_second": 1.029,
|
| 53 |
+
"eval_wer": 0.4126752336448598,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.5,
|
| 58 |
+
"learning_rate": 7.980000000000002e-06,
|
| 59 |
+
"loss": 0.9919,
|
| 60 |
+
"step": 400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.5,
|
| 64 |
+
"eval_loss": 0.7921169996261597,
|
| 65 |
+
"eval_runtime": 33.9628,
|
| 66 |
+
"eval_samples_per_second": 8.362,
|
| 67 |
+
"eval_steps_per_second": 1.06,
|
| 68 |
+
"eval_wer": 0.3919392523364486,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.12,
|
| 73 |
+
"learning_rate": 9.980000000000001e-06,
|
| 74 |
+
"loss": 0.9493,
|
| 75 |
+
"step": 500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.12,
|
| 79 |
+
"eval_loss": 0.7675915360450745,
|
| 80 |
+
"eval_runtime": 34.6559,
|
| 81 |
+
"eval_samples_per_second": 8.195,
|
| 82 |
+
"eval_steps_per_second": 1.039,
|
| 83 |
+
"eval_wer": 0.3744158878504673,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 3.75,
|
| 88 |
+
"learning_rate": 9.346666666666666e-06,
|
| 89 |
+
"loss": 0.9414,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 3.75,
|
| 94 |
+
"eval_loss": 0.7247262597084045,
|
| 95 |
+
"eval_runtime": 34.016,
|
| 96 |
+
"eval_samples_per_second": 8.349,
|
| 97 |
+
"eval_steps_per_second": 1.058,
|
| 98 |
+
"eval_wer": 0.3694509345794392,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.38,
|
| 103 |
+
"learning_rate": 8.68e-06,
|
| 104 |
+
"loss": 0.8897,
|
| 105 |
+
"step": 700
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 4.38,
|
| 109 |
+
"eval_loss": 0.720245897769928,
|
| 110 |
+
"eval_runtime": 35.0175,
|
| 111 |
+
"eval_samples_per_second": 8.11,
|
| 112 |
+
"eval_steps_per_second": 1.028,
|
| 113 |
+
"eval_wer": 0.3598130841121495,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.0,
|
| 118 |
+
"learning_rate": 8.013333333333333e-06,
|
| 119 |
+
"loss": 0.8716,
|
| 120 |
+
"step": 800
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.0,
|
| 124 |
+
"eval_loss": 0.7095639109611511,
|
| 125 |
+
"eval_runtime": 34.2057,
|
| 126 |
+
"eval_samples_per_second": 8.303,
|
| 127 |
+
"eval_steps_per_second": 1.052,
|
| 128 |
+
"eval_wer": 0.35455607476635514,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 5.62,
|
| 133 |
+
"learning_rate": 7.346666666666668e-06,
|
| 134 |
+
"loss": 0.8467,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 5.62,
|
| 139 |
+
"eval_loss": 0.7023197412490845,
|
| 140 |
+
"eval_runtime": 34.5651,
|
| 141 |
+
"eval_samples_per_second": 8.216,
|
| 142 |
+
"eval_steps_per_second": 1.042,
|
| 143 |
+
"eval_wer": 0.34988317757009346,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.25,
|
| 148 |
+
"learning_rate": 6.680000000000001e-06,
|
| 149 |
+
"loss": 0.8227,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.25,
|
| 154 |
+
"eval_loss": 0.6994471549987793,
|
| 155 |
+
"eval_runtime": 35.0644,
|
| 156 |
+
"eval_samples_per_second": 8.099,
|
| 157 |
+
"eval_steps_per_second": 1.027,
|
| 158 |
+
"eval_wer": 0.3411214953271028,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.88,
|
| 163 |
+
"learning_rate": 6.013333333333335e-06,
|
| 164 |
+
"loss": 0.855,
|
| 165 |
+
"step": 1100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 6.88,
|
| 169 |
+
"eval_loss": 0.6882991790771484,
|
| 170 |
+
"eval_runtime": 33.9338,
|
| 171 |
+
"eval_samples_per_second": 8.369,
|
| 172 |
+
"eval_steps_per_second": 1.061,
|
| 173 |
+
"eval_wer": 0.3431658878504673,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 7.5,
|
| 178 |
+
"learning_rate": 5.346666666666667e-06,
|
| 179 |
+
"loss": 0.8457,
|
| 180 |
+
"step": 1200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 7.5,
|
| 184 |
+
"eval_loss": 0.6772996187210083,
|
| 185 |
+
"eval_runtime": 33.7458,
|
| 186 |
+
"eval_samples_per_second": 8.416,
|
| 187 |
+
"eval_steps_per_second": 1.067,
|
| 188 |
+
"eval_wer": 0.34258177570093457,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.12,
|
| 193 |
+
"learning_rate": 4.680000000000001e-06,
|
| 194 |
+
"loss": 0.7614,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 8.12,
|
| 199 |
+
"eval_loss": 0.6912708282470703,
|
| 200 |
+
"eval_runtime": 33.9892,
|
| 201 |
+
"eval_samples_per_second": 8.356,
|
| 202 |
+
"eval_steps_per_second": 1.059,
|
| 203 |
+
"eval_wer": 0.33440420560747663,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 8.75,
|
| 208 |
+
"learning_rate": 4.013333333333334e-06,
|
| 209 |
+
"loss": 0.8127,
|
| 210 |
+
"step": 1400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 8.75,
|
| 214 |
+
"eval_loss": 0.6827006340026855,
|
| 215 |
+
"eval_runtime": 33.6998,
|
| 216 |
+
"eval_samples_per_second": 8.427,
|
| 217 |
+
"eval_steps_per_second": 1.068,
|
| 218 |
+
"eval_wer": 0.33352803738317754,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 9.38,
|
| 223 |
+
"learning_rate": 3.346666666666667e-06,
|
| 224 |
+
"loss": 0.8443,
|
| 225 |
+
"step": 1500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.38,
|
| 229 |
+
"eval_loss": 0.6724790930747986,
|
| 230 |
+
"eval_runtime": 34.0431,
|
| 231 |
+
"eval_samples_per_second": 8.342,
|
| 232 |
+
"eval_steps_per_second": 1.057,
|
| 233 |
+
"eval_wer": 0.33557242990654207,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"learning_rate": 2.68e-06,
|
| 239 |
+
"loss": 0.7548,
|
| 240 |
+
"step": 1600
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 10.0,
|
| 244 |
+
"eval_loss": 0.6758835315704346,
|
| 245 |
+
"eval_runtime": 34.9833,
|
| 246 |
+
"eval_samples_per_second": 8.118,
|
| 247 |
+
"eval_steps_per_second": 1.029,
|
| 248 |
+
"eval_wer": 0.3317757009345794,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 10.62,
|
| 253 |
+
"learning_rate": 2.0133333333333337e-06,
|
| 254 |
+
"loss": 0.7839,
|
| 255 |
+
"step": 1700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 10.62,
|
| 259 |
+
"eval_loss": 0.677266001701355,
|
| 260 |
+
"eval_runtime": 34.4604,
|
| 261 |
+
"eval_samples_per_second": 8.241,
|
| 262 |
+
"eval_steps_per_second": 1.045,
|
| 263 |
+
"eval_wer": 0.3285630841121495,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 11.25,
|
| 268 |
+
"learning_rate": 1.3466666666666668e-06,
|
| 269 |
+
"loss": 0.7912,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 11.25,
|
| 274 |
+
"eval_loss": 0.6747676730155945,
|
| 275 |
+
"eval_runtime": 35.0861,
|
| 276 |
+
"eval_samples_per_second": 8.094,
|
| 277 |
+
"eval_steps_per_second": 1.026,
|
| 278 |
+
"eval_wer": 0.3285630841121495,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 11.88,
|
| 283 |
+
"learning_rate": 6.800000000000001e-07,
|
| 284 |
+
"loss": 0.8238,
|
| 285 |
+
"step": 1900
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 11.88,
|
| 289 |
+
"eval_loss": 0.673457145690918,
|
| 290 |
+
"eval_runtime": 33.7389,
|
| 291 |
+
"eval_samples_per_second": 8.418,
|
| 292 |
+
"eval_steps_per_second": 1.067,
|
| 293 |
+
"eval_wer": 0.32973130841121495,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 12.5,
|
| 298 |
+
"learning_rate": 1.3333333333333334e-08,
|
| 299 |
+
"loss": 0.7618,
|
| 300 |
+
"step": 2000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 12.5,
|
| 304 |
+
"eval_loss": 0.672848641872406,
|
| 305 |
+
"eval_runtime": 34.5736,
|
| 306 |
+
"eval_samples_per_second": 8.214,
|
| 307 |
+
"eval_steps_per_second": 1.041,
|
| 308 |
+
"eval_wer": 0.3285630841121495,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 12.5,
|
| 313 |
+
"step": 2000,
|
| 314 |
+
"total_flos": 9.605345431463184e+18,
|
| 315 |
+
"train_loss": 0.9391711463928223,
|
| 316 |
+
"train_runtime": 9697.5717,
|
| 317 |
+
"train_samples_per_second": 3.3,
|
| 318 |
+
"train_steps_per_second": 0.206
|
| 319 |
+
}
|
| 320 |
+
],
|
| 321 |
+
"logging_steps": 100,
|
| 322 |
+
"max_steps": 2000,
|
| 323 |
+
"num_train_epochs": 13,
|
| 324 |
+
"save_steps": 100,
|
| 325 |
+
"total_flos": 9.605345431463184e+18,
|
| 326 |
+
"trial_name": null,
|
| 327 |
+
"trial_params": null
|
| 328 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6667d929fee99d07345e57d4876c1009c30b2915c37843db22a8d0f29919d030
|
| 3 |
+
size 4664
|
vocab.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<pad>": 0,
|
| 4 |
+
"<s>": 1,
|
| 5 |
+
"<unk>": 3,
|
| 6 |
+
"|": 4,
|
| 7 |
+
"آ": 5,
|
| 8 |
+
"ئ": 6,
|
| 9 |
+
"ا": 7,
|
| 10 |
+
"ب": 8,
|
| 11 |
+
"ت": 9,
|
| 12 |
+
"ث": 10,
|
| 13 |
+
"ج": 11,
|
| 14 |
+
"ح": 12,
|
| 15 |
+
"خ": 13,
|
| 16 |
+
"د": 14,
|
| 17 |
+
"ذ": 15,
|
| 18 |
+
"ر": 16,
|
| 19 |
+
"ز": 17,
|
| 20 |
+
"س": 18,
|
| 21 |
+
"ش": 19,
|
| 22 |
+
"ص": 20,
|
| 23 |
+
"ض": 21,
|
| 24 |
+
"ط": 22,
|
| 25 |
+
"ظ": 23,
|
| 26 |
+
"ع": 24,
|
| 27 |
+
"غ": 25,
|
| 28 |
+
"ف": 26,
|
| 29 |
+
"ق": 27,
|
| 30 |
+
"ل": 28,
|
| 31 |
+
"م": 29,
|
| 32 |
+
"ن": 30,
|
| 33 |
+
"ه": 31,
|
| 34 |
+
"و": 32,
|
| 35 |
+
"پ": 33,
|
| 36 |
+
"چ": 34,
|
| 37 |
+
"ژ": 35,
|
| 38 |
+
"ک": 36,
|
| 39 |
+
"گ": 37,
|
| 40 |
+
"ی": 38,
|
| 41 |
+
"": 39
|
| 42 |
+
}
|