roberta-base-finetuned-intent
This model is a fine-tuned version of roberta-base on the snips_built_in_intents dataset. It achieves the following results on the evaluation set:
- Loss: 0.2720
- Accuracy: 0.9333
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: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: IPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- total_eval_batch_size: 5
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- training precision: Mixed Precision
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.9568 | 1.0 | 37 | 1.7598 | 0.4333 |
1.2238 | 2.0 | 74 | 0.8130 | 0.7667 |
0.4536 | 3.0 | 111 | 0.4985 | 0.8 |
0.2478 | 4.0 | 148 | 0.3535 | 0.8667 |
0.0903 | 5.0 | 185 | 0.3110 | 0.8667 |
0.0849 | 6.0 | 222 | 0.2720 | 0.9333 |
0.0708 | 7.0 | 259 | 0.2742 | 0.8667 |
0.0796 | 8.0 | 296 | 0.2839 | 0.8667 |
0.0638 | 9.0 | 333 | 0.2949 | 0.8667 |
0.0566 | 10.0 | 370 | 0.2925 | 0.8667 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.10.0+cpu
- Datasets 2.7.1
- Tokenizers 0.12.0
- Downloads last month
- 17
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.