<!-- 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. -->
model_broadclass_onSet0try1
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9723
- 0 Precision: 0.7317
- 0 Recall: 0.9677
- 0 F1-score: 0.8333
- 0 Support: 31
- 1 Precision: 0.8276
- 1 Recall: 0.96
- 1 F1-score: 0.8889
- 1 Support: 25
- 2 Precision: 1.0
- 2 Recall: 0.7407
- 2 F1-score: 0.8511
- 2 Support: 27
- 3 Precision: 1.0
- 3 Recall: 0.5333
- 3 F1-score: 0.6957
- 3 Support: 15
- Accuracy: 0.8367
- Macro avg Precision: 0.8898
- Macro avg Recall: 0.8005
- Macro avg F1-score: 0.8172
- Macro avg Support: 98
- Weighted avg Precision: 0.8711
- Weighted avg Recall: 0.8367
- Weighted avg F1-score: 0.8313
- Weighted avg Support: 98
- Wer: 0.9220
- Mtrix: [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 1, 24, 0, 0], [2, 4, 3, 20, 0], [3, 6, 1, 0, 8]]
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 70
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | 0 Precision | 0 Recall | 0 F1-score | 0 Support | 1 Precision | 1 Recall | 1 F1-score | 1 Support | 2 Precision | 2 Recall | 2 F1-score | 2 Support | 3 Precision | 3 Recall | 3 F1-score | 3 Support | Accuracy | Macro avg Precision | Macro avg Recall | Macro avg F1-score | Macro avg Support | Weighted avg Precision | Weighted avg Recall | Weighted avg F1-score | Weighted avg Support | Wer | Mtrix |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2.329 | 4.16 | 100 | 2.2015 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
2.2772 | 8.33 | 200 | 2.1792 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
2.0617 | 12.49 | 300 | 2.0492 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.9607 | 16.65 | 400 | 1.8299 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.6665 | 20.82 | 500 | 1.5920 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.6451 | 24.98 | 600 | 1.5898 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.6024 | 29.16 | 700 | 1.5471 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.5967 | 33.33 | 800 | 1.5154 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
1.4451 | 37.49 | 900 | 1.4983 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
0.9896 | 41.65 | 1000 | 0.9953 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9842 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] |
0.9559 | 45.82 | 1100 | 0.9747 | 0.3483 | 1.0 | 0.5167 | 31 | 1.0 | 0.24 | 0.3871 | 25 | 1.0 | 0.0741 | 0.1379 | 27 | 1.0 | 0.0667 | 0.125 | 15 | 0.4082 | 0.8371 | 0.3452 | 0.2917 | 98 | 0.7939 | 0.4082 | 0.3193 | 98 | 0.9650 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 19, 6, 0, 0], [2, 25, 0, 2, 0], [3, 14, 0, 0, 1]] |
0.9441 | 49.98 | 1200 | 1.0000 | 0.4493 | 1.0 | 0.62 | 31 | 0.7857 | 0.44 | 0.5641 | 25 | 1.0 | 0.3333 | 0.5 | 27 | 1.0 | 0.4 | 0.5714 | 15 | 0.5816 | 0.8087 | 0.5433 | 0.5639 | 98 | 0.7711 | 0.5816 | 0.5652 | 98 | 0.9590 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 14, 11, 0, 0], [2, 15, 3, 9, 0], [3, 9, 0, 0, 6]] |
0.9656 | 54.16 | 1300 | 0.9814 | 0.5741 | 1.0 | 0.7294 | 31 | 0.8 | 0.64 | 0.7111 | 25 | 1.0 | 0.4444 | 0.6154 | 27 | 1.0 | 0.8 | 0.8889 | 15 | 0.7245 | 0.8435 | 0.7211 | 0.7362 | 98 | 0.8142 | 0.7245 | 0.7177 | 98 | 0.9304 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 9, 16, 0, 0], [2, 12, 3, 12, 0], [3, 2, 1, 0, 12]] |
0.9491 | 58.33 | 1400 | 0.9922 | 0.5 | 0.9677 | 0.6593 | 31 | 0.7778 | 0.56 | 0.6512 | 25 | 1.0 | 0.5185 | 0.6829 | 27 | 1.0 | 0.4 | 0.5714 | 15 | 0.6531 | 0.8194 | 0.6116 | 0.6412 | 98 | 0.7851 | 0.6531 | 0.6503 | 98 | 0.9383 | [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 11, 14, 0, 0], [2, 11, 2, 14, 0], [3, 8, 1, 0, 6]] |
0.8918 | 62.49 | 1500 | 0.9883 | 0.6522 | 0.9677 | 0.7792 | 31 | 0.8846 | 0.92 | 0.9020 | 25 | 1.0 | 0.5556 | 0.7143 | 27 | 1.0 | 0.7333 | 0.8462 | 15 | 0.8061 | 0.8842 | 0.7942 | 0.8104 | 98 | 0.8605 | 0.8061 | 0.8029 | 98 | 0.9383 | [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 2, 23, 0, 0], [2, 11, 1, 15, 0], [3, 3, 1, 0, 11]] |
0.8863 | 66.65 | 1600 | 0.9723 | 0.7317 | 0.9677 | 0.8333 | 31 | 0.8276 | 0.96 | 0.8889 | 25 | 1.0 | 0.7407 | 0.8511 | 27 | 1.0 | 0.5333 | 0.6957 | 15 | 0.8367 | 0.8898 | 0.8005 | 0.8172 | 98 | 0.8711 | 0.8367 | 0.8313 | 98 | 0.9220 | [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 1, 24, 0, 0], [2, 4, 3, 20, 0], [3, 6, 1, 0, 8]] |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2