|
---
|
|
license: apache-2.0
|
|
base_model: facebook/hubert-base-ls960
|
|
tags:
|
|
- generated_from_trainer
|
|
metrics:
|
|
- accuracy
|
|
- precision
|
|
- recall
|
|
- f1
|
|
model-index:
|
|
- name: hubert-classifier
|
|
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
|
|
|
|
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: 2.4637
|
|
- Accuracy: 0.5230
|
|
- Precision: 0.4945
|
|
- Recall: 0.5230
|
|
- F1: 0.4700
|
|
- Binary: 0.6634
|
|
|
|
## 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: 64
|
|
- eval_batch_size: 32
|
|
- seed: 42
|
|
- gradient_accumulation_steps: 4
|
|
- total_train_batch_size: 256
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
- lr_scheduler_type: linear
|
|
- num_epochs: 30
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary |
|
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:|
|
|
| No log | 1.72 | 50 | 4.2179 | 0.0484 | 0.0065 | 0.0484 | 0.0105 | 0.3058 |
|
|
| No log | 3.45 | 100 | 3.8319 | 0.1017 | 0.0846 | 0.1017 | 0.0618 | 0.3634 |
|
|
| No log | 5.17 | 150 | 3.5448 | 0.1864 | 0.1327 | 0.1864 | 0.1311 | 0.4274 |
|
|
| No log | 6.9 | 200 | 3.3129 | 0.2470 | 0.2063 | 0.2470 | 0.1855 | 0.4671 |
|
|
| No log | 8.62 | 250 | 3.1207 | 0.3123 | 0.3090 | 0.3123 | 0.2599 | 0.5150 |
|
|
| No log | 10.34 | 300 | 2.9535 | 0.3826 | 0.3524 | 0.3826 | 0.3277 | 0.5644 |
|
|
| No log | 12.07 | 350 | 2.8121 | 0.4310 | 0.3894 | 0.4310 | 0.3695 | 0.5983 |
|
|
| No log | 13.79 | 400 | 2.6726 | 0.4431 | 0.3939 | 0.4431 | 0.3775 | 0.6075 |
|
|
| No log | 15.52 | 450 | 2.5597 | 0.4818 | 0.4413 | 0.4818 | 0.4206 | 0.6370 |
|
|
| 3.4474 | 17.24 | 500 | 2.4637 | 0.5230 | 0.4945 | 0.5230 | 0.4700 | 0.6634 |
|
|
| 3.4474 | 18.97 | 550 | 2.3747 | 0.5400 | 0.5111 | 0.5400 | 0.4920 | 0.6760 |
|
|
| 3.4474 | 20.69 | 600 | 2.3113 | 0.5545 | 0.5212 | 0.5545 | 0.5067 | 0.6872 |
|
|
| 3.4474 | 22.41 | 650 | 2.2492 | 0.5714 | 0.5475 | 0.5714 | 0.5274 | 0.7007 |
|
|
| 3.4474 | 24.14 | 700 | 2.2053 | 0.5738 | 0.5511 | 0.5738 | 0.5336 | 0.7015 |
|
|
| 3.4474 | 25.86 | 750 | 2.1757 | 0.5714 | 0.5477 | 0.5714 | 0.5283 | 0.7015 |
|
|
| 3.4474 | 27.59 | 800 | 2.1491 | 0.5908 | 0.5574 | 0.5908 | 0.5468 | 0.7140 |
|
|
| 3.4474 | 29.31 | 850 | 2.1403 | 0.5932 | 0.5625 | 0.5932 | 0.5506 | 0.7167 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.38.2
|
|
- Pytorch 2.3.0
|
|
- Datasets 2.19.1
|
|
- Tokenizers 0.15.1
|
|
|
