<!-- 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. -->
bert_12_layer_model_v2_complete_training_new_72
This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_48 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.9589
- Accuracy: 0.4806
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: 1e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
3.6433 | 0.08 | 10000 | 3.5384 | 0.4110 |
3.5617 | 0.16 | 20000 | 3.4628 | 0.4191 |
3.489 | 0.25 | 30000 | 3.3906 | 0.4270 |
3.4379 | 0.33 | 40000 | 3.3331 | 0.4333 |
3.3763 | 0.41 | 50000 | 3.2776 | 0.4397 |
3.3329 | 0.49 | 60000 | 3.2282 | 0.4451 |
3.2794 | 0.57 | 70000 | 3.1804 | 0.4503 |
3.2344 | 0.66 | 80000 | 3.1357 | 0.4557 |
3.1767 | 0.74 | 90000 | 3.0819 | 0.4626 |
3.1205 | 0.82 | 100000 | 3.0137 | 0.4728 |
3.0641 | 0.9 | 110000 | 2.9589 | 0.4806 |
Framework versions
- Transformers 4.29.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3