|
---
|
|
license: apache-2.0
|
|
base_model: facebook/wav2vec2-base
|
|
tags:
|
|
- generated_from_trainer
|
|
metrics:
|
|
- accuracy
|
|
- precision
|
|
- recall
|
|
- f1
|
|
model-index:
|
|
- name: wav2vec2-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. -->
|
|
|
|
# wav2vec2-classifier-aug
|
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 0.9532
|
|
- Accuracy: 0.7951
|
|
- Precision: 0.8280
|
|
- Recall: 0.7951
|
|
- F1: 0.7944
|
|
- Binary: 0.8569
|
|
|
|
## 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: 3e-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 | 4.3438 | 0.0566 | 0.0357 | 0.0566 | 0.0270 | 0.3183 |
|
|
| No log | 0.38 | 100 | 3.9861 | 0.0916 | 0.0269 | 0.0916 | 0.0306 | 0.3604 |
|
|
| No log | 0.58 | 150 | 3.7184 | 0.1429 | 0.0837 | 0.1429 | 0.0727 | 0.3954 |
|
|
| No log | 0.77 | 200 | 3.5059 | 0.2345 | 0.2019 | 0.2345 | 0.1717 | 0.4631 |
|
|
| No log | 0.96 | 250 | 3.3078 | 0.2857 | 0.2031 | 0.2857 | 0.2069 | 0.4989 |
|
|
| 3.9492 | 1.15 | 300 | 3.1185 | 0.3396 | 0.2410 | 0.3396 | 0.2564 | 0.5412 |
|
|
| 3.9492 | 1.34 | 350 | 2.9671 | 0.3666 | 0.3559 | 0.3666 | 0.3003 | 0.5582 |
|
|
| 3.9492 | 1.53 | 400 | 2.8265 | 0.4151 | 0.3929 | 0.4151 | 0.3473 | 0.5911 |
|
|
| 3.9492 | 1.73 | 450 | 2.6754 | 0.5040 | 0.4403 | 0.5040 | 0.4386 | 0.6523 |
|
|
| 3.9492 | 1.92 | 500 | 2.5560 | 0.5148 | 0.4664 | 0.5148 | 0.4591 | 0.6593 |
|
|
| 3.0798 | 2.11 | 550 | 2.4304 | 0.5660 | 0.5781 | 0.5660 | 0.5243 | 0.6968 |
|
|
| 3.0798 | 2.3 | 600 | 2.3227 | 0.5768 | 0.5487 | 0.5768 | 0.5238 | 0.7065 |
|
|
| 3.0798 | 2.49 | 650 | 2.2025 | 0.6011 | 0.5784 | 0.6011 | 0.5567 | 0.7240 |
|
|
| 3.0798 | 2.68 | 700 | 2.1138 | 0.6199 | 0.5967 | 0.6199 | 0.5743 | 0.7326 |
|
|
| 3.0798 | 2.88 | 750 | 2.0058 | 0.6307 | 0.6042 | 0.6307 | 0.5840 | 0.7420 |
|
|
| 2.5472 | 3.07 | 800 | 1.9306 | 0.6577 | 0.6453 | 0.6577 | 0.6202 | 0.7617 |
|
|
| 2.5472 | 3.26 | 850 | 1.8359 | 0.6604 | 0.6700 | 0.6604 | 0.6243 | 0.7644 |
|
|
| 2.5472 | 3.45 | 900 | 1.7841 | 0.6712 | 0.6500 | 0.6712 | 0.6277 | 0.7701 |
|
|
| 2.5472 | 3.64 | 950 | 1.7083 | 0.6765 | 0.6906 | 0.6765 | 0.6423 | 0.7739 |
|
|
| 2.5472 | 3.84 | 1000 | 1.6311 | 0.7089 | 0.7518 | 0.7089 | 0.6864 | 0.7976 |
|
|
| 2.1755 | 4.03 | 1050 | 1.5739 | 0.6765 | 0.6801 | 0.6765 | 0.6404 | 0.7747 |
|
|
| 2.1755 | 4.22 | 1100 | 1.5318 | 0.7008 | 0.7197 | 0.7008 | 0.6679 | 0.7919 |
|
|
| 2.1755 | 4.41 | 1150 | 1.4939 | 0.7143 | 0.7365 | 0.7143 | 0.6908 | 0.8003 |
|
|
| 2.1755 | 4.6 | 1200 | 1.4532 | 0.7278 | 0.7410 | 0.7278 | 0.7051 | 0.8108 |
|
|
| 2.1755 | 4.79 | 1250 | 1.3933 | 0.7305 | 0.7554 | 0.7305 | 0.7152 | 0.8127 |
|
|
| 2.1755 | 4.99 | 1300 | 1.3863 | 0.7143 | 0.7404 | 0.7143 | 0.6923 | 0.8013 |
|
|
| 1.9226 | 5.18 | 1350 | 1.3064 | 0.7493 | 0.7955 | 0.7493 | 0.7351 | 0.8248 |
|
|
| 1.9226 | 5.37 | 1400 | 1.2828 | 0.7520 | 0.7651 | 0.7520 | 0.7361 | 0.8288 |
|
|
| 1.9226 | 5.56 | 1450 | 1.2408 | 0.7520 | 0.7739 | 0.7520 | 0.7374 | 0.8267 |
|
|
| 1.9226 | 5.75 | 1500 | 1.2134 | 0.7628 | 0.7862 | 0.7628 | 0.7507 | 0.8342 |
|
|
| 1.9226 | 5.94 | 1550 | 1.1905 | 0.7628 | 0.7948 | 0.7628 | 0.7484 | 0.8353 |
|
|
| 1.7422 | 6.14 | 1600 | 1.1820 | 0.7547 | 0.7966 | 0.7547 | 0.7427 | 0.8286 |
|
|
| 1.7422 | 6.33 | 1650 | 1.1576 | 0.7574 | 0.8034 | 0.7574 | 0.7453 | 0.8305 |
|
|
| 1.7422 | 6.52 | 1700 | 1.1313 | 0.7574 | 0.7991 | 0.7574 | 0.7486 | 0.8315 |
|
|
| 1.7422 | 6.71 | 1750 | 1.1140 | 0.7709 | 0.8030 | 0.7709 | 0.7620 | 0.8410 |
|
|
| 1.7422 | 6.9 | 1800 | 1.0973 | 0.7628 | 0.7881 | 0.7628 | 0.7563 | 0.8353 |
|
|
| 1.6131 | 7.09 | 1850 | 1.0891 | 0.7709 | 0.8084 | 0.7709 | 0.7670 | 0.8412 |
|
|
| 1.6131 | 7.29 | 1900 | 1.0634 | 0.7601 | 0.7970 | 0.7601 | 0.7520 | 0.8345 |
|
|
| 1.6131 | 7.48 | 1950 | 1.0521 | 0.7736 | 0.7937 | 0.7736 | 0.7655 | 0.8439 |
|
|
| 1.6131 | 7.67 | 2000 | 1.0334 | 0.7817 | 0.8189 | 0.7817 | 0.7752 | 0.8474 |
|
|
| 1.6131 | 7.86 | 2050 | 1.0233 | 0.7844 | 0.8077 | 0.7844 | 0.7740 | 0.8504 |
|
|
| 1.5198 | 8.05 | 2100 | 1.0091 | 0.7817 | 0.8098 | 0.7817 | 0.7735 | 0.8474 |
|
|
| 1.5198 | 8.25 | 2150 | 1.0165 | 0.7709 | 0.8120 | 0.7709 | 0.7669 | 0.8399 |
|
|
| 1.5198 | 8.44 | 2200 | 0.9963 | 0.7790 | 0.8136 | 0.7790 | 0.7721 | 0.8456 |
|
|
| 1.5198 | 8.63 | 2250 | 0.9857 | 0.7763 | 0.8197 | 0.7763 | 0.7728 | 0.8437 |
|
|
| 1.5198 | 8.82 | 2300 | 0.9730 | 0.7898 | 0.8255 | 0.7898 | 0.7877 | 0.8531 |
|
|
| 1.4558 | 9.01 | 2350 | 0.9699 | 0.7978 | 0.8347 | 0.7978 | 0.7965 | 0.8588 |
|
|
| 1.4558 | 9.2 | 2400 | 0.9636 | 0.7925 | 0.8309 | 0.7925 | 0.7927 | 0.8550 |
|
|
| 1.4558 | 9.4 | 2450 | 0.9541 | 0.7898 | 0.8252 | 0.7898 | 0.7891 | 0.8531 |
|
|
| 1.4558 | 9.59 | 2500 | 0.9534 | 0.7925 | 0.8317 | 0.7925 | 0.7939 | 0.8550 |
|
|
| 1.4558 | 9.78 | 2550 | 0.9518 | 0.7951 | 0.8280 | 0.7951 | 0.7944 | 0.8569 |
|
|
| 1.4558 | 9.97 | 2600 | 0.9532 | 0.7951 | 0.8280 | 0.7951 | 0.7944 | 0.8569 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.38.2
|
|
- Pytorch 2.3.0
|
|
- Datasets 2.19.1
|
|
- Tokenizers 0.15.1
|
|
|
