mbart_jokes

This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0282

Model description

This model is trained of jokes dataset , where you can ask a question and the model gives funny answer.

Intended uses & limitations

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
3.3455	1.0	1914	3.0282

Framework versions

• Transformers 4.12.5
• Pytorch 1.9.1
• Datasets 1.16.1
• Tokenizers 0.10.3