cuenb

This model is a version of roberta-base with adaptive pre-training using the negation-focused pre-training strategy on 1.2M sentences containing negation.

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: 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: 5.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.8998	2.57	200000	1.6606

Framework versions

• Transformers 4.13.0.dev0
• Pytorch 1.8.0
• Datasets 1.15.1
• Tokenizers 0.10.3