|
---
|
|
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-fold-2
|
|
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-fold-2
|
|
|
|
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: 0.7259
|
|
- Accuracy: 0.8720
|
|
- Precision: 0.8869
|
|
- Recall: 0.8720
|
|
- F1: 0.8705
|
|
- Binary: 0.9115
|
|
|
|
## 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: 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
|
|
- lr_scheduler_warmup_steps: 500
|
|
- num_epochs: 30
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Binary |
|
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:|
|
|
| No log | 0.13 | 50 | 4.4197 | 0.0270 | 0.0232 | 0.0270 | 0.0117 | 0.1849 |
|
|
| No log | 0.27 | 100 | 4.2998 | 0.0432 | 0.0232 | 0.0432 | 0.0109 | 0.3126 |
|
|
| No log | 0.4 | 150 | 3.9731 | 0.0553 | 0.0154 | 0.0553 | 0.0170 | 0.3332 |
|
|
| No log | 0.54 | 200 | 3.6492 | 0.0594 | 0.0211 | 0.0594 | 0.0169 | 0.3355 |
|
|
| No log | 0.67 | 250 | 3.4482 | 0.0945 | 0.0252 | 0.0945 | 0.0317 | 0.3599 |
|
|
| No log | 0.81 | 300 | 3.2094 | 0.1215 | 0.0559 | 0.1215 | 0.0495 | 0.3815 |
|
|
| No log | 0.94 | 350 | 3.0184 | 0.1660 | 0.1000 | 0.1660 | 0.0883 | 0.4140 |
|
|
| 3.8426 | 1.08 | 400 | 2.6834 | 0.2807 | 0.2104 | 0.2807 | 0.2054 | 0.4961 |
|
|
| 3.8426 | 1.21 | 450 | 2.3799 | 0.3900 | 0.3073 | 0.3900 | 0.3055 | 0.5725 |
|
|
| 3.8426 | 1.35 | 500 | 2.0607 | 0.4440 | 0.4235 | 0.4440 | 0.3765 | 0.6111 |
|
|
| 3.8426 | 1.48 | 550 | 1.7232 | 0.5520 | 0.5014 | 0.5520 | 0.4917 | 0.6857 |
|
|
| 3.8426 | 1.62 | 600 | 1.5077 | 0.6019 | 0.5531 | 0.6019 | 0.5435 | 0.7232 |
|
|
| 3.8426 | 1.75 | 650 | 1.3166 | 0.6181 | 0.6024 | 0.6181 | 0.5743 | 0.7335 |
|
|
| 3.8426 | 1.89 | 700 | 1.2097 | 0.6802 | 0.6583 | 0.6802 | 0.6449 | 0.7750 |
|
|
| 2.1078 | 2.02 | 750 | 1.1776 | 0.6856 | 0.6934 | 0.6856 | 0.6510 | 0.7808 |
|
|
| 2.1078 | 2.16 | 800 | 1.0102 | 0.7382 | 0.7335 | 0.7382 | 0.7132 | 0.8152 |
|
|
| 2.1078 | 2.29 | 850 | 0.9295 | 0.7409 | 0.7557 | 0.7409 | 0.7242 | 0.8189 |
|
|
| 2.1078 | 2.43 | 900 | 0.8473 | 0.7760 | 0.7916 | 0.7760 | 0.7673 | 0.8444 |
|
|
| 2.1078 | 2.56 | 950 | 0.7956 | 0.8057 | 0.8147 | 0.8057 | 0.7962 | 0.8632 |
|
|
| 2.1078 | 2.7 | 1000 | 0.8091 | 0.7908 | 0.8199 | 0.7908 | 0.7817 | 0.8534 |
|
|
| 2.1078 | 2.83 | 1050 | 0.7345 | 0.8084 | 0.8208 | 0.8084 | 0.8027 | 0.8650 |
|
|
| 2.1078 | 2.96 | 1100 | 0.7217 | 0.7935 | 0.7964 | 0.7935 | 0.7835 | 0.8557 |
|
|
| 1.1182 | 3.1 | 1150 | 0.7444 | 0.7895 | 0.8096 | 0.7895 | 0.7827 | 0.8545 |
|
|
| 1.1182 | 3.23 | 1200 | 0.7182 | 0.8057 | 0.8153 | 0.8057 | 0.7971 | 0.8638 |
|
|
| 1.1182 | 3.37 | 1250 | 0.6496 | 0.8313 | 0.8414 | 0.8313 | 0.8267 | 0.8808 |
|
|
| 1.1182 | 3.5 | 1300 | 0.7163 | 0.8111 | 0.8349 | 0.8111 | 0.8101 | 0.8684 |
|
|
| 1.1182 | 3.64 | 1350 | 0.7026 | 0.8354 | 0.8543 | 0.8354 | 0.8309 | 0.8846 |
|
|
| 1.1182 | 3.77 | 1400 | 0.6504 | 0.8246 | 0.8381 | 0.8246 | 0.8194 | 0.8765 |
|
|
| 1.1182 | 3.91 | 1450 | 0.6685 | 0.8381 | 0.8525 | 0.8381 | 0.8358 | 0.8879 |
|
|
| 0.7752 | 4.04 | 1500 | 0.6600 | 0.8340 | 0.8528 | 0.8340 | 0.8314 | 0.8839 |
|
|
| 0.7752 | 4.18 | 1550 | 0.6048 | 0.8475 | 0.8594 | 0.8475 | 0.8440 | 0.8920 |
|
|
| 0.7752 | 4.31 | 1600 | 0.5907 | 0.8435 | 0.8575 | 0.8435 | 0.8397 | 0.8906 |
|
|
| 0.7752 | 4.45 | 1650 | 0.6118 | 0.8502 | 0.8673 | 0.8502 | 0.8476 | 0.8949 |
|
|
| 0.7752 | 4.58 | 1700 | 0.6096 | 0.8610 | 0.8724 | 0.8610 | 0.8581 | 0.9024 |
|
|
| 0.7752 | 4.72 | 1750 | 0.6032 | 0.8529 | 0.8694 | 0.8529 | 0.8499 | 0.8977 |
|
|
| 0.7752 | 4.85 | 1800 | 0.6705 | 0.8502 | 0.8625 | 0.8502 | 0.8447 | 0.8962 |
|
|
| 0.7752 | 4.99 | 1850 | 0.6740 | 0.8381 | 0.8538 | 0.8381 | 0.8331 | 0.8877 |
|
|
| 0.6054 | 5.12 | 1900 | 0.6444 | 0.8367 | 0.8471 | 0.8367 | 0.8302 | 0.8853 |
|
|
| 0.6054 | 5.26 | 1950 | 0.6167 | 0.8529 | 0.8648 | 0.8529 | 0.8500 | 0.8976 |
|
|
| 0.6054 | 5.39 | 2000 | 0.6535 | 0.8462 | 0.8657 | 0.8462 | 0.8429 | 0.8939 |
|
|
| 0.6054 | 5.53 | 2050 | 0.6420 | 0.8556 | 0.8688 | 0.8556 | 0.8550 | 0.8992 |
|
|
| 0.6054 | 5.66 | 2100 | 0.6535 | 0.8543 | 0.8696 | 0.8543 | 0.8523 | 0.8977 |
|
|
| 0.6054 | 5.8 | 2150 | 0.5879 | 0.8516 | 0.8646 | 0.8516 | 0.8483 | 0.8958 |
|
|
| 0.6054 | 5.93 | 2200 | 0.5808 | 0.8570 | 0.8701 | 0.8570 | 0.8551 | 0.9000 |
|
|
| 0.5025 | 6.06 | 2250 | 0.6637 | 0.8516 | 0.8711 | 0.8516 | 0.8482 | 0.8962 |
|
|
| 0.5025 | 6.2 | 2300 | 0.6450 | 0.8583 | 0.8734 | 0.8583 | 0.8540 | 0.9009 |
|
|
| 0.5025 | 6.33 | 2350 | 0.6152 | 0.8664 | 0.8768 | 0.8664 | 0.8644 | 0.9082 |
|
|
| 0.5025 | 6.47 | 2400 | 0.6640 | 0.8475 | 0.8620 | 0.8475 | 0.8432 | 0.8934 |
|
|
| 0.5025 | 6.6 | 2450 | 0.5817 | 0.8664 | 0.8795 | 0.8664 | 0.8645 | 0.9062 |
|
|
| 0.5025 | 6.74 | 2500 | 0.6881 | 0.8529 | 0.8673 | 0.8529 | 0.8487 | 0.8977 |
|
|
| 0.5025 | 6.87 | 2550 | 0.6868 | 0.8421 | 0.8563 | 0.8421 | 0.8388 | 0.8907 |
|
|
| 0.4381 | 7.01 | 2600 | 0.6270 | 0.8677 | 0.8823 | 0.8677 | 0.8664 | 0.9086 |
|
|
| 0.4381 | 7.14 | 2650 | 0.7011 | 0.8583 | 0.8703 | 0.8583 | 0.8537 | 0.9001 |
|
|
| 0.4381 | 7.28 | 2700 | 0.6665 | 0.8570 | 0.8757 | 0.8570 | 0.8548 | 0.8992 |
|
|
| 0.4381 | 7.41 | 2750 | 0.6948 | 0.8421 | 0.8586 | 0.8421 | 0.8425 | 0.8911 |
|
|
| 0.4381 | 7.55 | 2800 | 0.6832 | 0.8570 | 0.8710 | 0.8570 | 0.8542 | 0.9005 |
|
|
| 0.4381 | 7.68 | 2850 | 0.6391 | 0.8623 | 0.8782 | 0.8623 | 0.8620 | 0.9038 |
|
|
| 0.4381 | 7.82 | 2900 | 0.8113 | 0.8448 | 0.8609 | 0.8448 | 0.8411 | 0.8946 |
|
|
| 0.4381 | 7.95 | 2950 | 0.6688 | 0.8623 | 0.8724 | 0.8623 | 0.8603 | 0.9049 |
|
|
| 0.381 | 8.09 | 3000 | 0.6731 | 0.8529 | 0.8652 | 0.8529 | 0.8508 | 0.8972 |
|
|
| 0.381 | 8.22 | 3050 | 0.8063 | 0.8340 | 0.8507 | 0.8340 | 0.8300 | 0.8839 |
|
|
| 0.381 | 8.36 | 3100 | 0.6534 | 0.8596 | 0.8719 | 0.8596 | 0.8567 | 0.9009 |
|
|
| 0.381 | 8.49 | 3150 | 0.6772 | 0.8596 | 0.8730 | 0.8596 | 0.8574 | 0.9024 |
|
|
| 0.381 | 8.63 | 3200 | 0.6293 | 0.8637 | 0.8754 | 0.8637 | 0.8630 | 0.9042 |
|
|
| 0.381 | 8.76 | 3250 | 0.6644 | 0.8570 | 0.8705 | 0.8570 | 0.8543 | 0.9001 |
|
|
| 0.381 | 8.89 | 3300 | 0.6469 | 0.8623 | 0.8758 | 0.8623 | 0.8614 | 0.9043 |
|
|
| 0.3473 | 9.03 | 3350 | 0.6055 | 0.8731 | 0.8833 | 0.8731 | 0.8722 | 0.9104 |
|
|
| 0.3473 | 9.16 | 3400 | 0.6828 | 0.8650 | 0.8767 | 0.8650 | 0.8636 | 0.9047 |
|
|
| 0.3473 | 9.3 | 3450 | 0.6625 | 0.8826 | 0.8967 | 0.8826 | 0.8817 | 0.9179 |
|
|
| 0.3473 | 9.43 | 3500 | 0.7111 | 0.8583 | 0.8692 | 0.8583 | 0.8559 | 0.9011 |
|
|
| 0.3473 | 9.57 | 3550 | 0.7215 | 0.8475 | 0.8608 | 0.8475 | 0.8449 | 0.8950 |
|
|
| 0.3473 | 9.7 | 3600 | 0.7040 | 0.8556 | 0.8654 | 0.8556 | 0.8527 | 0.9001 |
|
|
| 0.3473 | 9.84 | 3650 | 0.6809 | 0.8556 | 0.8674 | 0.8556 | 0.8531 | 0.8996 |
|
|
| 0.3473 | 9.97 | 3700 | 0.7191 | 0.8610 | 0.8754 | 0.8610 | 0.8609 | 0.9034 |
|
|
| 0.3245 | 10.11 | 3750 | 0.7053 | 0.8610 | 0.8682 | 0.8610 | 0.8586 | 0.9045 |
|
|
| 0.3245 | 10.24 | 3800 | 0.6594 | 0.8745 | 0.8869 | 0.8745 | 0.8739 | 0.9132 |
|
|
| 0.3245 | 10.38 | 3850 | 0.6882 | 0.8745 | 0.8872 | 0.8745 | 0.8735 | 0.9113 |
|
|
| 0.3245 | 10.51 | 3900 | 0.7113 | 0.8596 | 0.8732 | 0.8596 | 0.8584 | 0.9035 |
|
|
| 0.3245 | 10.65 | 3950 | 0.7299 | 0.8677 | 0.8836 | 0.8677 | 0.8675 | 0.9070 |
|
|
| 0.3245 | 10.78 | 4000 | 0.6812 | 0.8758 | 0.8861 | 0.8758 | 0.8745 | 0.9132 |
|
|
| 0.3245 | 10.92 | 4050 | 0.6459 | 0.8812 | 0.8927 | 0.8812 | 0.8788 | 0.9170 |
|
|
| 0.2964 | 11.05 | 4100 | 0.7044 | 0.8677 | 0.8805 | 0.8677 | 0.8665 | 0.9072 |
|
|
| 0.2964 | 11.19 | 4150 | 0.6455 | 0.8677 | 0.8758 | 0.8677 | 0.8663 | 0.9086 |
|
|
| 0.2964 | 11.32 | 4200 | 0.7581 | 0.8704 | 0.8790 | 0.8704 | 0.8692 | 0.9093 |
|
|
| 0.2964 | 11.46 | 4250 | 0.7489 | 0.8623 | 0.8781 | 0.8623 | 0.8588 | 0.9038 |
|
|
| 0.2964 | 11.59 | 4300 | 0.7293 | 0.8556 | 0.8677 | 0.8556 | 0.8541 | 0.8991 |
|
|
| 0.2964 | 11.73 | 4350 | 0.7996 | 0.8570 | 0.8662 | 0.8570 | 0.8550 | 0.8991 |
|
|
| 0.2964 | 11.86 | 4400 | 0.7340 | 0.8556 | 0.8670 | 0.8556 | 0.8531 | 0.8985 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.38.2
|
|
- Pytorch 2.3.0
|
|
- Datasets 2.19.1
|
|
- Tokenizers 0.15.1
|
|
|
