<!-- 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. -->
mtg_1000_subset_wav2vec2_model
This model is a fine-tuned version of facebook/wav2vec2-base-100k-voxpopuli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.3077
- Accuracy: 0.09
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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.96 | 6 | 2.3026 | 0.11 |
| 2.3035 | 1.92 | 12 | 2.3044 | 0.08 |
| 2.3035 | 2.88 | 18 | 2.3056 | 0.075 |
| 2.3027 | 4.0 | 25 | 2.3064 | 0.08 |
| 2.302 | 4.96 | 31 | 2.3070 | 0.08 |
| 2.302 | 5.92 | 37 | 2.3072 | 0.09 |
| 2.3006 | 6.88 | 43 | 2.3075 | 0.09 |
| 2.3022 | 8.0 | 50 | 2.3077 | 0.09 |
| 2.3022 | 8.96 | 56 | 2.3077 | 0.09 |
| 2.3025 | 9.6 | 60 | 2.3077 | 0.09 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
