File size: 5,354 Bytes
36642a6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: kids_phoneme_sm_model
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# kids_phoneme_sm_model
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.4558
- Cer: 0.4079
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 3.0642 | 0.74 | 500 | 4.4995 | 1.0 |
| 2.8486 | 1.48 | 1000 | 3.8639 | 1.0 |
| 2.7909 | 2.22 | 1500 | 3.4712 | 1.0 |
| 1.5475 | 2.96 | 2000 | 1.0263 | 0.6825 |
| 0.7353 | 3.7 | 2500 | 0.8291 | 0.5760 |
| 0.6036 | 4.44 | 3000 | 0.7387 | 0.5327 |
| 0.5553 | 5.19 | 3500 | 0.7382 | 0.5023 |
| 0.4271 | 5.93 | 4000 | 0.7244 | 0.4991 |
| 0.43 | 6.67 | 4500 | 0.7152 | 0.4805 |
| 0.3925 | 7.41 | 5000 | 0.7210 | 0.4587 |
| 0.3719 | 8.15 | 5500 | 0.7888 | 0.4491 |
| 0.3451 | 8.89 | 6000 | 0.7599 | 0.4433 |
| 0.319 | 9.63 | 6500 | 0.7642 | 0.4508 |
| 0.2638 | 10.37 | 7000 | 0.8490 | 0.4426 |
| 0.3084 | 11.11 | 7500 | 0.9387 | 0.4315 |
| 0.2553 | 11.85 | 8000 | 0.8477 | 0.4287 |
| 0.2537 | 12.59 | 8500 | 0.8261 | 0.4301 |
| 0.2058 | 13.33 | 9000 | 1.1093 | 0.4247 |
| 0.2283 | 14.07 | 9500 | 0.7638 | 0.4230 |
| 0.2043 | 14.81 | 10000 | 1.0104 | 0.4219 |
| 0.1918 | 15.56 | 10500 | 0.9618 | 0.4194 |
| 0.1764 | 16.3 | 11000 | 0.9460 | 0.4226 |
| 0.1677 | 17.04 | 11500 | 0.9750 | 0.4233 |
| 0.1751 | 17.78 | 12000 | 0.9600 | 0.4240 |
| 0.1465 | 18.52 | 12500 | 1.1328 | 0.4172 |
| 0.1239 | 19.26 | 13000 | 1.0746 | 0.4176 |
| 0.1495 | 20.0 | 13500 | 1.2143 | 0.4194 |
| 0.1444 | 20.74 | 14000 | 1.1595 | 0.4219 |
| 0.134 | 21.48 | 14500 | 1.1601 | 0.4201 |
| 0.1343 | 22.22 | 15000 | 1.1730 | 0.4233 |
| 0.1051 | 22.96 | 15500 | 1.1257 | 0.4172 |
| 0.1067 | 23.7 | 16000 | 1.1206 | 0.4190 |
| 0.0959 | 24.44 | 16500 | 1.1539 | 0.4133 |
| 0.1028 | 25.19 | 17000 | 1.2425 | 0.4126 |
| 0.1028 | 25.93 | 17500 | 1.2008 | 0.4144 |
| 0.1052 | 26.67 | 18000 | 1.1974 | 0.4094 |
| 0.0813 | 27.41 | 18500 | 1.0960 | 0.4133 |
| 0.0973 | 28.15 | 19000 | 1.1153 | 0.4101 |
| 0.0783 | 28.89 | 19500 | 1.1596 | 0.4126 |
| 0.0704 | 29.63 | 20000 | 1.1881 | 0.4087 |
| 0.068 | 30.37 | 20500 | 1.2289 | 0.4040 |
| 0.0664 | 31.11 | 21000 | 1.2289 | 0.4079 |
| 0.0747 | 31.85 | 21500 | 1.2642 | 0.4122 |
| 0.0663 | 32.59 | 22000 | 1.3062 | 0.4101 |
| 0.0668 | 33.33 | 22500 | 1.3486 | 0.4101 |
| 0.0592 | 34.07 | 23000 | 1.3346 | 0.4040 |
| 0.0513 | 34.81 | 23500 | 1.2958 | 0.4097 |
| 0.0511 | 35.56 | 24000 | 1.3798 | 0.4108 |
| 0.0557 | 36.3 | 24500 | 1.3521 | 0.4065 |
| 0.049 | 37.04 | 25000 | 1.4192 | 0.4094 |
| 0.0465 | 37.78 | 25500 | 1.4308 | 0.4108 |
| 0.0474 | 38.52 | 26000 | 1.4004 | 0.4058 |
| 0.0428 | 39.26 | 26500 | 1.3988 | 0.4054 |
| 0.0509 | 40.0 | 27000 | 1.4218 | 0.4069 |
| 0.0386 | 40.74 | 27500 | 1.3819 | 0.4104 |
| 0.0426 | 41.48 | 28000 | 1.4681 | 0.4090 |
| 0.0408 | 42.22 | 28500 | 1.4543 | 0.4104 |
| 0.0405 | 42.96 | 29000 | 1.4999 | 0.4108 |
| 0.036 | 43.7 | 29500 | 1.4922 | 0.4072 |
| 0.036 | 44.44 | 30000 | 1.4709 | 0.4087 |
| 0.04 | 45.19 | 30500 | 1.4858 | 0.4094 |
| 0.0343 | 45.93 | 31000 | 1.4606 | 0.4087 |
| 0.0288 | 46.67 | 31500 | 1.4599 | 0.4044 |
| 0.0454 | 47.41 | 32000 | 1.4288 | 0.4087 |
| 0.0322 | 48.15 | 32500 | 1.4589 | 0.4083 |
| 0.0327 | 48.89 | 33000 | 1.4502 | 0.4094 |
| 0.0272 | 49.63 | 33500 | 1.4558 | 0.4079 |
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.0
- Datasets 2.12.0
- Tokenizers 0.13.3
|