File size: 6,972 Bytes
cc6c2d2 9089351 cc6c2d2 9089351 cc6c2d2 9089351 cc6c2d2 9089351 878d563 cc6c2d2 878d563 cc6c2d2 878d563 cc6c2d2 9089351 cc6c2d2 |
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 |
---
license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: hubert-classifier-aug
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. -->
# hubert-classifier-aug
This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.0783
- Accuracy: 0.2075
- Precision: 0.1563
- Recall: 0.2075
- F1: 0.1504
- Binary: 0.4396
## 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: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:|
| No log | 0.19 | 50 | 4.4148 | 0.0243 | 0.0244 | 0.0243 | 0.0114 | 0.1755 |
| No log | 0.38 | 100 | 4.3445 | 0.0404 | 0.0101 | 0.0404 | 0.0123 | 0.2865 |
| No log | 0.58 | 150 | 4.2100 | 0.0350 | 0.0154 | 0.0350 | 0.0080 | 0.3043 |
| No log | 0.77 | 200 | 4.1002 | 0.0350 | 0.0185 | 0.0350 | 0.0088 | 0.3140 |
| No log | 0.96 | 250 | 4.0141 | 0.0512 | 0.0295 | 0.0512 | 0.0232 | 0.3253 |
| No log | 1.15 | 300 | 3.9438 | 0.0593 | 0.0321 | 0.0593 | 0.0293 | 0.3318 |
| No log | 1.34 | 350 | 3.8728 | 0.0647 | 0.0290 | 0.0647 | 0.0281 | 0.3372 |
| No log | 1.53 | 400 | 3.8297 | 0.0755 | 0.0242 | 0.0755 | 0.0331 | 0.3423 |
| No log | 1.73 | 450 | 3.7627 | 0.0620 | 0.0186 | 0.0620 | 0.0263 | 0.3385 |
| 4.147 | 1.92 | 500 | 3.7166 | 0.0728 | 0.0364 | 0.0728 | 0.0360 | 0.3437 |
| 4.147 | 2.11 | 550 | 3.6779 | 0.0889 | 0.0425 | 0.0889 | 0.0486 | 0.3558 |
| 4.147 | 2.3 | 600 | 3.6396 | 0.0755 | 0.0345 | 0.0755 | 0.0407 | 0.3447 |
| 4.147 | 2.49 | 650 | 3.6005 | 0.0889 | 0.0336 | 0.0889 | 0.0412 | 0.3550 |
| 4.147 | 2.68 | 700 | 3.5602 | 0.0970 | 0.0314 | 0.0970 | 0.0420 | 0.3631 |
| 4.147 | 2.88 | 750 | 3.5309 | 0.0997 | 0.0473 | 0.0997 | 0.0507 | 0.3642 |
| 4.147 | 3.07 | 800 | 3.5331 | 0.1051 | 0.0385 | 0.1051 | 0.0490 | 0.3615 |
| 4.147 | 3.26 | 850 | 3.4774 | 0.1105 | 0.0507 | 0.1105 | 0.0604 | 0.3701 |
| 4.147 | 3.45 | 900 | 3.4571 | 0.1159 | 0.0568 | 0.1159 | 0.0611 | 0.3730 |
| 4.147 | 3.64 | 950 | 3.4265 | 0.1132 | 0.0431 | 0.1132 | 0.0582 | 0.3736 |
| 3.6862 | 3.84 | 1000 | 3.4260 | 0.0970 | 0.0406 | 0.0970 | 0.0502 | 0.3582 |
| 3.6862 | 4.03 | 1050 | 3.3821 | 0.1105 | 0.0421 | 0.1105 | 0.0542 | 0.3709 |
| 3.6862 | 4.22 | 1100 | 3.3825 | 0.1186 | 0.0448 | 0.1186 | 0.0578 | 0.3725 |
| 3.6862 | 4.41 | 1150 | 3.3575 | 0.1213 | 0.0507 | 0.1213 | 0.0634 | 0.3776 |
| 3.6862 | 4.6 | 1200 | 3.3453 | 0.1267 | 0.0659 | 0.1267 | 0.0653 | 0.3790 |
| 3.6862 | 4.79 | 1250 | 3.3205 | 0.1321 | 0.0592 | 0.1321 | 0.0736 | 0.3871 |
| 3.6862 | 4.99 | 1300 | 3.2912 | 0.1294 | 0.0552 | 0.1294 | 0.0724 | 0.3868 |
| 3.6862 | 5.18 | 1350 | 3.2741 | 0.1536 | 0.0731 | 0.1536 | 0.0880 | 0.4022 |
| 3.6862 | 5.37 | 1400 | 3.2767 | 0.1509 | 0.0723 | 0.1509 | 0.0893 | 0.3978 |
| 3.6862 | 5.56 | 1450 | 3.2485 | 0.1509 | 0.0743 | 0.1509 | 0.0907 | 0.4003 |
| 3.4619 | 5.75 | 1500 | 3.2421 | 0.1509 | 0.0783 | 0.1509 | 0.0855 | 0.4003 |
| 3.4619 | 5.94 | 1550 | 3.2366 | 0.1375 | 0.0686 | 0.1375 | 0.0754 | 0.3892 |
| 3.4619 | 6.14 | 1600 | 3.2102 | 0.1456 | 0.0959 | 0.1456 | 0.0862 | 0.3965 |
| 3.4619 | 6.33 | 1650 | 3.1962 | 0.1456 | 0.0688 | 0.1456 | 0.0858 | 0.3957 |
| 3.4619 | 6.52 | 1700 | 3.1917 | 0.1590 | 0.1160 | 0.1590 | 0.0994 | 0.4035 |
| 3.4619 | 6.71 | 1750 | 3.1746 | 0.1590 | 0.0922 | 0.1590 | 0.0978 | 0.4051 |
| 3.4619 | 6.9 | 1800 | 3.1791 | 0.1590 | 0.0671 | 0.1590 | 0.0863 | 0.4059 |
| 3.4619 | 7.09 | 1850 | 3.1714 | 0.1725 | 0.0952 | 0.1725 | 0.1028 | 0.4135 |
| 3.4619 | 7.29 | 1900 | 3.1427 | 0.1725 | 0.1084 | 0.1725 | 0.1090 | 0.4194 |
| 3.4619 | 7.48 | 1950 | 3.1410 | 0.1833 | 0.1313 | 0.1833 | 0.1221 | 0.4226 |
| 3.3361 | 7.67 | 2000 | 3.1334 | 0.1806 | 0.1385 | 0.1806 | 0.1239 | 0.4197 |
| 3.3361 | 7.86 | 2050 | 3.1246 | 0.1806 | 0.1474 | 0.1806 | 0.1193 | 0.4208 |
| 3.3361 | 8.05 | 2100 | 3.1151 | 0.1995 | 0.1582 | 0.1995 | 0.1388 | 0.4332 |
| 3.3361 | 8.25 | 2150 | 3.1085 | 0.2049 | 0.1578 | 0.2049 | 0.1439 | 0.4377 |
| 3.3361 | 8.44 | 2200 | 3.0897 | 0.2102 | 0.1546 | 0.2102 | 0.1483 | 0.4445 |
| 3.3361 | 8.63 | 2250 | 3.0934 | 0.2210 | 0.1511 | 0.2210 | 0.1541 | 0.4469 |
| 3.3361 | 8.82 | 2300 | 3.0906 | 0.2102 | 0.1625 | 0.2102 | 0.1535 | 0.4394 |
| 3.3361 | 9.01 | 2350 | 3.0792 | 0.2129 | 0.1586 | 0.2129 | 0.1573 | 0.4437 |
| 3.3361 | 9.2 | 2400 | 3.0849 | 0.2049 | 0.1442 | 0.2049 | 0.1446 | 0.4358 |
| 3.3361 | 9.4 | 2450 | 3.0794 | 0.2102 | 0.1576 | 0.2102 | 0.1532 | 0.4396 |
| 3.2647 | 9.59 | 2500 | 3.0801 | 0.2129 | 0.1560 | 0.2129 | 0.1552 | 0.4415 |
| 3.2647 | 9.78 | 2550 | 3.0823 | 0.2075 | 0.1669 | 0.2075 | 0.1521 | 0.4396 |
| 3.2647 | 9.97 | 2600 | 3.0783 | 0.2075 | 0.1563 | 0.2075 | 0.1504 | 0.4396 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.15.1
|