bert_12_layer_model_v1_complete_training_new_96
This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_72 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.3874
- Accuracy: 0.5637
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 |
---|---|---|---|---|
2.6463 | 0.08 | 10000 | 2.5710 | 0.5405 |
2.6202 | 0.16 | 20000 | 2.5489 | 0.5435 |
2.593 | 0.25 | 30000 | 2.5222 | 0.5465 |
2.5802 | 0.33 | 40000 | 2.4992 | 0.5495 |
2.5477 | 0.41 | 50000 | 2.4760 | 0.5525 |
2.5316 | 0.49 | 60000 | 2.4572 | 0.5548 |
2.504 | 0.57 | 70000 | 2.4368 | 0.5572 |
2.4878 | 0.66 | 80000 | 2.4225 | 0.5592 |
2.4667 | 0.74 | 90000 | 2.4007 | 0.5616 |
2.4513 | 0.82 | 100000 | 2.3874 | 0.5637 |
Framework versions
- Transformers 4.30.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 7
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.