Commit
•
9b2efc8
1
Parent(s):
ee39f68
End of training
Browse files- README.md +160 -0
- config.json +109 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- trainer_state.json +343 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,160 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
base_model: facebook/mms-300m
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: mms-300m-sah
|
| 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 |
+
# mms-300m-sah
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3105
|
| 21 |
+
- Wer: 0.3059
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 32
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- lr_scheduler_warmup_steps: 1000
|
| 47 |
+
- num_epochs: 100
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 53 |
+
| 8.3074 | 1.0 | 111 | 4.1552 | 1.0 |
|
| 54 |
+
| 3.6543 | 2.0 | 222 | 3.2635 | 1.0 |
|
| 55 |
+
| 3.109 | 3.0 | 333 | 2.9604 | 1.0 |
|
| 56 |
+
| 2.221 | 4.0 | 444 | 0.9272 | 0.7549 |
|
| 57 |
+
| 0.6842 | 5.0 | 555 | 0.4823 | 0.5726 |
|
| 58 |
+
| 0.4123 | 6.0 | 666 | 0.3828 | 0.5006 |
|
| 59 |
+
| 0.3021 | 7.0 | 777 | 0.3563 | 0.4868 |
|
| 60 |
+
| 0.2589 | 8.0 | 888 | 0.3188 | 0.4482 |
|
| 61 |
+
| 0.2246 | 9.0 | 999 | 0.3108 | 0.4430 |
|
| 62 |
+
| 0.1896 | 10.0 | 1110 | 0.3100 | 0.4130 |
|
| 63 |
+
| 0.1695 | 11.0 | 1221 | 0.2926 | 0.4104 |
|
| 64 |
+
| 0.1528 | 12.0 | 1332 | 0.2906 | 0.4133 |
|
| 65 |
+
| 0.1385 | 13.0 | 1443 | 0.2815 | 0.3931 |
|
| 66 |
+
| 0.1267 | 14.0 | 1554 | 0.3070 | 0.3966 |
|
| 67 |
+
| 0.1194 | 15.0 | 1665 | 0.2917 | 0.3877 |
|
| 68 |
+
| 0.1102 | 16.0 | 1776 | 0.2896 | 0.3805 |
|
| 69 |
+
| 0.1056 | 17.0 | 1887 | 0.2768 | 0.3793 |
|
| 70 |
+
| 0.099 | 18.0 | 1998 | 0.2910 | 0.3782 |
|
| 71 |
+
| 0.0897 | 19.0 | 2109 | 0.3145 | 0.3793 |
|
| 72 |
+
| 0.0876 | 20.0 | 2220 | 0.3028 | 0.3710 |
|
| 73 |
+
| 0.0878 | 21.0 | 2331 | 0.2956 | 0.3744 |
|
| 74 |
+
| 0.0877 | 22.0 | 2442 | 0.2894 | 0.3730 |
|
| 75 |
+
| 0.0851 | 23.0 | 2553 | 0.3086 | 0.3805 |
|
| 76 |
+
| 0.0825 | 24.0 | 2664 | 0.3168 | 0.3744 |
|
| 77 |
+
| 0.0765 | 25.0 | 2775 | 0.3113 | 0.3615 |
|
| 78 |
+
| 0.0778 | 26.0 | 2886 | 0.3204 | 0.3744 |
|
| 79 |
+
| 0.0777 | 27.0 | 2997 | 0.3257 | 0.3727 |
|
| 80 |
+
| 0.0752 | 28.0 | 3108 | 0.3118 | 0.3612 |
|
| 81 |
+
| 0.0736 | 29.0 | 3219 | 0.3159 | 0.3638 |
|
| 82 |
+
| 0.0677 | 30.0 | 3330 | 0.2975 | 0.3540 |
|
| 83 |
+
| 0.0663 | 31.0 | 3441 | 0.3080 | 0.3548 |
|
| 84 |
+
| 0.0655 | 32.0 | 3552 | 0.3223 | 0.3597 |
|
| 85 |
+
| 0.0658 | 33.0 | 3663 | 0.3215 | 0.3571 |
|
| 86 |
+
| 0.0664 | 34.0 | 3774 | 0.3164 | 0.3733 |
|
| 87 |
+
| 0.0635 | 35.0 | 3885 | 0.3239 | 0.3586 |
|
| 88 |
+
| 0.0621 | 36.0 | 3996 | 0.3188 | 0.3586 |
|
| 89 |
+
| 0.06 | 37.0 | 4107 | 0.2937 | 0.3563 |
|
| 90 |
+
| 0.0572 | 38.0 | 4218 | 0.3262 | 0.3620 |
|
| 91 |
+
| 0.0576 | 39.0 | 4329 | 0.3097 | 0.3505 |
|
| 92 |
+
| 0.0571 | 40.0 | 4440 | 0.3086 | 0.3580 |
|
| 93 |
+
| 0.0559 | 41.0 | 4551 | 0.3257 | 0.3641 |
|
| 94 |
+
| 0.0581 | 42.0 | 4662 | 0.3245 | 0.3537 |
|
| 95 |
+
| 0.0542 | 43.0 | 4773 | 0.3193 | 0.3612 |
|
| 96 |
+
| 0.0516 | 44.0 | 4884 | 0.2950 | 0.3531 |
|
| 97 |
+
| 0.0553 | 45.0 | 4995 | 0.3261 | 0.3522 |
|
| 98 |
+
| 0.0508 | 46.0 | 5106 | 0.3347 | 0.3563 |
|
| 99 |
+
| 0.0478 | 47.0 | 5217 | 0.3229 | 0.3600 |
|
| 100 |
+
| 0.0468 | 48.0 | 5328 | 0.3134 | 0.3482 |
|
| 101 |
+
| 0.0478 | 49.0 | 5439 | 0.3087 | 0.3491 |
|
| 102 |
+
| 0.045 | 50.0 | 5550 | 0.3103 | 0.3361 |
|
| 103 |
+
| 0.0485 | 51.0 | 5661 | 0.3148 | 0.3476 |
|
| 104 |
+
| 0.0438 | 52.0 | 5772 | 0.3138 | 0.3448 |
|
| 105 |
+
| 0.0444 | 53.0 | 5883 | 0.3151 | 0.3407 |
|
| 106 |
+
| 0.0447 | 54.0 | 5994 | 0.2992 | 0.3355 |
|
| 107 |
+
| 0.0439 | 55.0 | 6105 | 0.3165 | 0.3436 |
|
| 108 |
+
| 0.0413 | 56.0 | 6216 | 0.3184 | 0.3384 |
|
| 109 |
+
| 0.0394 | 57.0 | 6327 | 0.3217 | 0.3404 |
|
| 110 |
+
| 0.0413 | 58.0 | 6438 | 0.3062 | 0.3315 |
|
| 111 |
+
| 0.0386 | 59.0 | 6549 | 0.2985 | 0.3255 |
|
| 112 |
+
| 0.039 | 60.0 | 6660 | 0.3125 | 0.3407 |
|
| 113 |
+
| 0.038 | 61.0 | 6771 | 0.2937 | 0.3381 |
|
| 114 |
+
| 0.0361 | 62.0 | 6882 | 0.3138 | 0.3318 |
|
| 115 |
+
| 0.0359 | 63.0 | 6993 | 0.3296 | 0.3315 |
|
| 116 |
+
| 0.0347 | 64.0 | 7104 | 0.3260 | 0.3355 |
|
| 117 |
+
| 0.036 | 65.0 | 7215 | 0.3003 | 0.3373 |
|
| 118 |
+
| 0.0366 | 66.0 | 7326 | 0.2967 | 0.3283 |
|
| 119 |
+
| 0.0321 | 67.0 | 7437 | 0.3035 | 0.3240 |
|
| 120 |
+
| 0.0308 | 68.0 | 7548 | 0.3335 | 0.3390 |
|
| 121 |
+
| 0.0311 | 69.0 | 7659 | 0.3096 | 0.3263 |
|
| 122 |
+
| 0.0325 | 70.0 | 7770 | 0.3164 | 0.3306 |
|
| 123 |
+
| 0.032 | 71.0 | 7881 | 0.2890 | 0.3211 |
|
| 124 |
+
| 0.0312 | 72.0 | 7992 | 0.2847 | 0.3194 |
|
| 125 |
+
| 0.0289 | 73.0 | 8103 | 0.2904 | 0.3200 |
|
| 126 |
+
| 0.0289 | 74.0 | 8214 | 0.2932 | 0.3174 |
|
| 127 |
+
| 0.0276 | 75.0 | 8325 | 0.2921 | 0.3168 |
|
| 128 |
+
| 0.0277 | 76.0 | 8436 | 0.3054 | 0.3200 |
|
| 129 |
+
| 0.0271 | 77.0 | 8547 | 0.3078 | 0.3197 |
|
| 130 |
+
| 0.0261 | 78.0 | 8658 | 0.3191 | 0.3220 |
|
| 131 |
+
| 0.0268 | 79.0 | 8769 | 0.3081 | 0.3211 |
|
| 132 |
+
| 0.0251 | 80.0 | 8880 | 0.3089 | 0.3142 |
|
| 133 |
+
| 0.0245 | 81.0 | 8991 | 0.3081 | 0.3151 |
|
| 134 |
+
| 0.0229 | 82.0 | 9102 | 0.3124 | 0.3148 |
|
| 135 |
+
| 0.0232 | 83.0 | 9213 | 0.3074 | 0.3142 |
|
| 136 |
+
| 0.0241 | 84.0 | 9324 | 0.3045 | 0.3111 |
|
| 137 |
+
| 0.0213 | 85.0 | 9435 | 0.3234 | 0.3131 |
|
| 138 |
+
| 0.0215 | 86.0 | 9546 | 0.3148 | 0.3105 |
|
| 139 |
+
| 0.0209 | 87.0 | 9657 | 0.3160 | 0.3134 |
|
| 140 |
+
| 0.0208 | 88.0 | 9768 | 0.3055 | 0.3099 |
|
| 141 |
+
| 0.0201 | 89.0 | 9879 | 0.2996 | 0.3065 |
|
| 142 |
+
| 0.0196 | 90.0 | 9990 | 0.3036 | 0.3073 |
|
| 143 |
+
| 0.0187 | 91.0 | 10101 | 0.3137 | 0.3111 |
|
| 144 |
+
| 0.0189 | 92.0 | 10212 | 0.3089 | 0.3067 |
|
| 145 |
+
| 0.0184 | 93.0 | 10323 | 0.3118 | 0.3113 |
|
| 146 |
+
| 0.0172 | 94.0 | 10434 | 0.3081 | 0.3105 |
|
| 147 |
+
| 0.018 | 95.0 | 10545 | 0.3108 | 0.3099 |
|
| 148 |
+
| 0.0164 | 96.0 | 10656 | 0.3081 | 0.3073 |
|
| 149 |
+
| 0.0175 | 97.0 | 10767 | 0.3100 | 0.3082 |
|
| 150 |
+
| 0.0159 | 98.0 | 10878 | 0.3124 | 0.3056 |
|
| 151 |
+
| 0.0181 | 99.0 | 10989 | 0.3093 | 0.3044 |
|
| 152 |
+
| 0.0161 | 100.0 | 11100 | 0.3105 | 0.3059 |
|
| 153 |
+
|
| 154 |
+
|
| 155 |
+
### Framework versions
|
| 156 |
+
|
| 157 |
+
- Transformers 4.32.0
|
| 158 |
+
- Pytorch 2.0.1+cu117
|
| 159 |
+
- Datasets 2.14.4
|
| 160 |
+
- Tokenizers 0.13.3
|
config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mms-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": 0.0,
|
| 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": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 4096,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.0,
|
| 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": 39,
|
| 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": "4.32.0",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 40,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:163c7c9e4ad7bcb69223a02ca376b336b2600a60c7e61ecb65b45b00f0205e04
|
| 3 |
+
size 1262065837
|
trainer_state.json
ADDED
|
@@ -0,0 +1,343 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 75.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.57,
|
| 13 |
+
"learning_rate": 2.5e-05,
|
| 14 |
+
"loss": 8.1262,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 3.57,
|
| 19 |
+
"eval_loss": 4.0765790939331055,
|
| 20 |
+
"eval_runtime": 21.8341,
|
| 21 |
+
"eval_samples_per_second": 18.091,
|
| 22 |
+
"eval_steps_per_second": 0.595,
|
| 23 |
+
"eval_wer": 1.0,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 7.14,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 3.5989,
|
| 30 |
+
"step": 200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 7.14,
|
| 34 |
+
"eval_loss": 3.226550340652466,
|
| 35 |
+
"eval_runtime": 21.5714,
|
| 36 |
+
"eval_samples_per_second": 18.311,
|
| 37 |
+
"eval_steps_per_second": 0.603,
|
| 38 |
+
"eval_wer": 1.0,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 10.71,
|
| 43 |
+
"learning_rate": 7.5e-05,
|
| 44 |
+
"loss": 3.0472,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 10.71,
|
| 49 |
+
"eval_loss": 2.8889501094818115,
|
| 50 |
+
"eval_runtime": 21.7358,
|
| 51 |
+
"eval_samples_per_second": 18.173,
|
| 52 |
+
"eval_steps_per_second": 0.598,
|
| 53 |
+
"eval_wer": 1.0,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 14.29,
|
| 58 |
+
"learning_rate": 0.0001,
|
| 59 |
+
"loss": 1.7327,
|
| 60 |
+
"step": 400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 14.29,
|
| 64 |
+
"eval_loss": 0.6693879961967468,
|
| 65 |
+
"eval_runtime": 22.0147,
|
| 66 |
+
"eval_samples_per_second": 17.943,
|
| 67 |
+
"eval_steps_per_second": 0.591,
|
| 68 |
+
"eval_wer": 0.6622969506982046,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 17.86,
|
| 73 |
+
"learning_rate": 0.000125,
|
| 74 |
+
"loss": 0.4239,
|
| 75 |
+
"step": 500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 17.86,
|
| 79 |
+
"eval_loss": 0.4327137768268585,
|
| 80 |
+
"eval_runtime": 21.6438,
|
| 81 |
+
"eval_samples_per_second": 18.25,
|
| 82 |
+
"eval_steps_per_second": 0.601,
|
| 83 |
+
"eval_wer": 0.5483043602165859,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 21.43,
|
| 88 |
+
"learning_rate": 0.00015,
|
| 89 |
+
"loss": 0.2337,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 21.43,
|
| 94 |
+
"eval_loss": 0.38920775055885315,
|
| 95 |
+
"eval_runtime": 21.8228,
|
| 96 |
+
"eval_samples_per_second": 18.1,
|
| 97 |
+
"eval_steps_per_second": 0.596,
|
| 98 |
+
"eval_wer": 0.5195212311199772,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 25.0,
|
| 103 |
+
"learning_rate": 0.000175,
|
| 104 |
+
"loss": 0.1724,
|
| 105 |
+
"step": 700
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 25.0,
|
| 109 |
+
"eval_loss": 0.391787052154541,
|
| 110 |
+
"eval_runtime": 21.5976,
|
| 111 |
+
"eval_samples_per_second": 18.289,
|
| 112 |
+
"eval_steps_per_second": 0.602,
|
| 113 |
+
"eval_wer": 0.5027073240239385,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 28.57,
|
| 118 |
+
"learning_rate": 0.0002,
|
| 119 |
+
"loss": 0.1392,
|
| 120 |
+
"step": 800
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 28.57,
|
| 124 |
+
"eval_loss": 0.36863410472869873,
|
| 125 |
+
"eval_runtime": 21.5216,
|
| 126 |
+
"eval_samples_per_second": 18.354,
|
| 127 |
+
"eval_steps_per_second": 0.604,
|
| 128 |
+
"eval_wer": 0.4670846394984326,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 32.14,
|
| 133 |
+
"learning_rate": 0.00022500000000000002,
|
| 134 |
+
"loss": 0.1196,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 32.14,
|
| 139 |
+
"eval_loss": 0.3849872052669525,
|
| 140 |
+
"eval_runtime": 21.7525,
|
| 141 |
+
"eval_samples_per_second": 18.159,
|
| 142 |
+
"eval_steps_per_second": 0.598,
|
| 143 |
+
"eval_wer": 0.45397549159304645,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 35.71,
|
| 148 |
+
"learning_rate": 0.00025,
|
| 149 |
+
"loss": 0.1095,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 35.71,
|
| 154 |
+
"eval_loss": 0.3586702346801758,
|
| 155 |
+
"eval_runtime": 21.8949,
|
| 156 |
+
"eval_samples_per_second": 18.041,
|
| 157 |
+
"eval_steps_per_second": 0.594,
|
| 158 |
+
"eval_wer": 0.4516956397834141,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 39.29,
|
| 163 |
+
"learning_rate": 0.000275,
|
| 164 |
+
"loss": 0.1026,
|
| 165 |
+
"step": 1100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 39.29,
|
| 169 |
+
"eval_loss": 0.36062636971473694,
|
| 170 |
+
"eval_runtime": 21.8096,
|
| 171 |
+
"eval_samples_per_second": 18.111,
|
| 172 |
+
"eval_steps_per_second": 0.596,
|
| 173 |
+
"eval_wer": 0.44029638073525224,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 42.86,
|
| 178 |
+
"learning_rate": 0.0003,
|
| 179 |
+
"loss": 0.0972,
|
| 180 |
+
"step": 1200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 42.86,
|
| 184 |
+
"eval_loss": 0.40814533829689026,
|
| 185 |
+
"eval_runtime": 21.9118,
|
| 186 |
+
"eval_samples_per_second": 18.027,
|
| 187 |
+
"eval_steps_per_second": 0.593,
|
| 188 |
+
"eval_wer": 0.43317184383015106,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 46.43,
|
| 193 |
+
"learning_rate": 0.00032500000000000004,
|
| 194 |
+
"loss": 0.0932,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 46.43,
|
| 199 |
+
"eval_loss": 0.3904629945755005,
|
| 200 |
+
"eval_runtime": 21.7505,
|
| 201 |
+
"eval_samples_per_second": 18.16,
|
| 202 |
+
"eval_steps_per_second": 0.598,
|
| 203 |
+
"eval_wer": 0.426902251353662,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 50.0,
|
| 208 |
+
"learning_rate": 0.00035,
|
| 209 |
+
"loss": 0.0897,
|
| 210 |
+
"step": 1400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 50.0,
|
| 214 |
+
"eval_loss": 0.37547701597213745,
|
| 215 |
+
"eval_runtime": 22.1102,
|
| 216 |
+
"eval_samples_per_second": 17.865,
|
| 217 |
+
"eval_steps_per_second": 0.588,
|
| 218 |
+
"eval_wer": 0.42747221430607013,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 53.57,
|
| 223 |
+
"learning_rate": 0.000375,
|
| 224 |
+
"loss": 0.0846,
|
| 225 |
+
"step": 1500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 53.57,
|
| 229 |
+
"eval_loss": 0.36823779344558716,
|
| 230 |
+
"eval_runtime": 21.896,
|
| 231 |
+
"eval_samples_per_second": 18.04,
|
| 232 |
+
"eval_steps_per_second": 0.594,
|
| 233 |
+
"eval_wer": 0.42091764035337703,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 57.14,
|
| 238 |
+
"learning_rate": 0.0004,
|
| 239 |
+
"loss": 0.0854,
|
| 240 |
+
"step": 1600
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 57.14,
|
| 244 |
+
"eval_loss": 0.3795730471611023,
|
| 245 |
+
"eval_runtime": 21.5795,
|
| 246 |
+
"eval_samples_per_second": 18.304,
|
| 247 |
+
"eval_steps_per_second": 0.602,
|
| 248 |
+
"eval_wer": 0.41635793673411226,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 60.71,
|
| 253 |
+
"learning_rate": 0.000425,
|
| 254 |
+
"loss": 0.0845,
|
| 255 |
+
"step": 1700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 60.71,
|
| 259 |
+
"eval_loss": 0.35857513546943665,
|
| 260 |
+
"eval_runtime": 21.7949,
|
| 261 |
+
"eval_samples_per_second": 18.124,
|
| 262 |
+
"eval_steps_per_second": 0.596,
|
| 263 |
+
"eval_wer": 0.3941293815901966,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 64.29,
|
| 268 |
+
"learning_rate": 0.00045000000000000004,
|
| 269 |
+
"loss": 0.0854,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 64.29,
|
| 274 |
+
"eval_loss": 0.376005083322525,
|
| 275 |
+
"eval_runtime": 21.5391,
|
| 276 |
+
"eval_samples_per_second": 18.339,
|
| 277 |
+
"eval_steps_per_second": 0.604,
|
| 278 |
+
"eval_wer": 0.41664291821031635,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 67.86,
|
| 283 |
+
"learning_rate": 0.000475,
|
| 284 |
+
"loss": 0.0846,
|
| 285 |
+
"step": 1900
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 67.86,
|
| 289 |
+
"eval_loss": 0.3710671663284302,
|
| 290 |
+
"eval_runtime": 21.5817,
|
| 291 |
+
"eval_samples_per_second": 18.303,
|
| 292 |
+
"eval_steps_per_second": 0.602,
|
| 293 |
+
"eval_wer": 0.41208321459105157,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 71.43,
|
| 298 |
+
"learning_rate": 0.0005,
|
| 299 |
+
"loss": 0.0827,
|
| 300 |
+
"step": 2000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 71.43,
|
| 304 |
+
"eval_loss": 0.40105244517326355,
|
| 305 |
+
"eval_runtime": 21.6021,
|
| 306 |
+
"eval_samples_per_second": 18.285,
|
| 307 |
+
"eval_steps_per_second": 0.602,
|
| 308 |
+
"eval_wer": 0.4160729552579082,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 75.0,
|
| 313 |
+
"learning_rate": 0.0,
|
| 314 |
+
"loss": 0.0665,
|
| 315 |
+
"step": 2100
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 75.0,
|
| 319 |
+
"eval_loss": 0.3478299379348755,
|
| 320 |
+
"eval_runtime": 21.8418,
|
| 321 |
+
"eval_samples_per_second": 18.085,
|
| 322 |
+
"eval_steps_per_second": 0.595,
|
| 323 |
+
"eval_wer": 0.36534625249358793,
|
| 324 |
+
"step": 2100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 75.0,
|
| 328 |
+
"step": 2100,
|
| 329 |
+
"total_flos": 5.419723804611084e+19,
|
| 330 |
+
"train_loss": 0.8885498528253465,
|
| 331 |
+
"train_runtime": 26470.8294,
|
| 332 |
+
"train_samples_per_second": 10.064,
|
| 333 |
+
"train_steps_per_second": 0.079
|
| 334 |
+
}
|
| 335 |
+
],
|
| 336 |
+
"logging_steps": 100,
|
| 337 |
+
"max_steps": 2100,
|
| 338 |
+
"num_train_epochs": 75,
|
| 339 |
+
"save_steps": 100,
|
| 340 |
+
"total_flos": 5.419723804611084e+19,
|
| 341 |
+
"trial_name": null,
|
| 342 |
+
"trial_params": null
|
| 343 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1df3cce6090eaa854b099c6d231b661e82235cd17328262b79311c041363a021
|
| 3 |
+
size 4027
|