hbertv1-tiny-wt-48-Massive-intent-emb-comp
This model is a fine-tuned version of gokuls/model_v1_complete_training_wt_init_48_tiny_emb_comp on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.8545
- Accuracy: 0.7900
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 3.6847 | 1.0 | 180 | 3.2207 | 0.2710 |
| 2.7795 | 2.0 | 360 | 2.3154 | 0.4471 |
| 2.0459 | 3.0 | 540 | 1.7680 | 0.5627 |
| 1.5874 | 4.0 | 720 | 1.4363 | 0.6734 |
| 1.2902 | 5.0 | 900 | 1.2306 | 0.7127 |
| 1.0905 | 6.0 | 1080 | 1.1068 | 0.7373 |
| 0.9468 | 7.0 | 1260 | 1.0113 | 0.7545 |
| 0.844 | 8.0 | 1440 | 0.9661 | 0.7580 |
| 0.7684 | 9.0 | 1620 | 0.9333 | 0.7649 |
| 0.7086 | 10.0 | 1800 | 0.9018 | 0.7772 |
| 0.6629 | 11.0 | 1980 | 0.8807 | 0.7831 |
| 0.6244 | 12.0 | 2160 | 0.8747 | 0.7796 |
| 0.5965 | 13.0 | 2340 | 0.8591 | 0.7875 |
| 0.5731 | 14.0 | 2520 | 0.8634 | 0.7875 |
| 0.5633 | 15.0 | 2700 | 0.8545 | 0.7900 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3