This model is a fine-tuned version of t5-small on MultiWOZ 2.1 both user and system utterances.
Refer to ConvLab-3 for model description and usage.
The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 128
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 256
- optimizer: Adafactor
- lr_scheduler_type: linear
- num_epochs: 10.0
- Transformers 4.20.1
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
- Downloads last month
This model can be loaded on the Inference API on-demand.
Dataset used to train ConvLab/t5-small-nlu-all-multiwoz21
- Accuracy on MultiWOZ 2.1test set self-reported69.100
- F1 on MultiWOZ 2.1test set self-reported83.500