metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: bert_12_layer_model_v1_complete_training_new_wt_init_120
results: []
bert_12_layer_model_v1_complete_training_new_wt_init_120
This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_96 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.1966
- Accuracy: 0.5856
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 |
|---|---|---|---|---|
| 2.3673 | 0.08 | 10000 | 2.2852 | 0.5732 |
| 2.356 | 0.16 | 20000 | 2.2772 | 0.5744 |
| 2.3424 | 0.25 | 30000 | 2.2640 | 0.5765 |
| 2.3442 | 0.33 | 40000 | 2.2525 | 0.5778 |
| 2.3228 | 0.41 | 50000 | 2.2427 | 0.5793 |
| 2.3179 | 0.49 | 60000 | 2.2313 | 0.5810 |
| 2.2993 | 0.57 | 70000 | 2.2237 | 0.5822 |
| 2.2911 | 0.66 | 80000 | 2.2128 | 0.5831 |
| 2.279 | 0.74 | 90000 | 2.2008 | 0.5842 |
| 2.2715 | 0.82 | 100000 | 2.1966 | 0.5856 |
Framework versions
- Transformers 4.30.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3