roberta-base-culinary

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

• Loss: 1.1032

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss
1.5135	1.0	39823	1.4635
1.454	2.0	79646	1.3753
1.3924	3.0	119469	1.3375
1.3379	4.0	159292	1.2886
1.2969	5.0	199115	1.2595
1.2495	6.0	238938	nan
1.1768	7.0	278761	1.2283
1.1687	8.0	318584	1.2109
1.2148	9.0	358407	1.1671
1.133	10.0	398230	1.1721
1.0882	11.0	438053	1.1624
1.0749	12.0	477876	1.1321
1.092	13.0	517699	nan
1.0594	14.0	557522	1.1186
1.0292	15.0	597345	1.1074
0.9973	16.0	637168	1.1032

Framework versions

• Transformers 4.22.1
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1