metadata
license: mit
tags:
- generated_from_trainer
datasets:
- snips_built_in_intents
metrics:
- accuracy
model-index:
- name: roberta-base-finetuned-intent
results: []
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: 0.2158
- Accuracy: 1.0
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 |
|---|---|---|---|---|
| 1.7528 | 1.0 | 37 | 1.5430 | 0.4 |
| 0.4361 | 2.0 | 74 | 0.6562 | 0.8667 |
| 0.2656 | 3.0 | 111 | 0.2158 | 1.0 |
| 0.1121 | 4.0 | 148 | 0.1056 | 1.0 |
| 0.1295 | 5.0 | 185 | 0.0753 | 1.0 |
| 0.0897 | 6.0 | 222 | 0.0554 | 1.0 |
| 0.0704 | 7.0 | 259 | 0.0467 | 1.0 |
| 0.0638 | 8.0 | 296 | 0.0420 | 1.0 |
| 0.047 | 9.0 | 333 | 0.0393 | 1.0 |
| 0.0632 | 10.0 | 370 | 0.0387 | 1.0 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.10.0+cpu
- Datasets 2.7.1
- Tokenizers 0.12.0