Edit model card

recipe-roberta-upper-Is

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

  • Loss: 0.7757

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: 256
  • eval_batch_size: 256
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
1.2455 1.0 1228 1.0420
1.0812 2.0 2456 0.9641
1.018 3.0 3684 0.9220
0.977 4.0 4912 0.8943
0.9451 5.0 6140 0.8726
0.9254 6.0 7368 0.8574
0.9074 7.0 8596 0.8404
0.8944 8.0 9824 0.8290
0.8797 9.0 11052 0.8258
0.869 10.0 12280 0.8115
0.8609 11.0 13508 0.8085
0.8522 12.0 14736 0.7995
0.8462 13.0 15964 0.7958
0.8414 14.0 17192 0.7891
0.8374 15.0 18420 0.7856
0.8327 16.0 19648 0.7850
0.8268 17.0 20876 0.7784
0.8256 18.0 22104 0.7802
0.822 19.0 23332 0.7789
0.8219 20.0 24560 0.7757

Framework versions

  • Transformers 4.19.0.dev0
  • Pytorch 1.11.0+cu102
  • Datasets 2.3.2
  • Tokenizers 0.12.1
Downloads last month
13