recipe-roberta-upper-tIs

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

• Loss: 0.7904

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
1.2671	1.0	1281	1.0554
1.0995	2.0	2562	0.9832
1.0339	3.0	3843	0.9389
0.9925	4.0	5124	0.9095
0.964	5.0	6405	0.8914
0.9426	6.0	7686	0.8708
0.9227	7.0	8967	0.8590
0.9082	8.0	10248	0.8448
0.8963	9.0	11529	0.8361
0.8847	10.0	12810	0.8249
0.8756	11.0	14091	0.8204
0.8672	12.0	15372	0.8105
0.8612	13.0	16653	0.8106
0.8561	14.0	17934	0.8041
0.8485	15.0	19215	0.7979
0.8452	16.0	20496	0.7910
0.8403	17.0	21777	0.7991
0.8389	18.0	23058	0.7928
0.8371	19.0	24339	0.7926
0.8341	20.0	25620	0.7904

Framework versions

• Transformers 4.19.0.dev0
• Pytorch 1.11.0+cu102
• Datasets 2.3.2
• Tokenizers 0.12.1