recipe-roberta-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: 0.8491

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
1.3552	1.0	1012	1.1292
1.1811	2.0	2024	1.0543
1.1095	3.0	3036	1.0122
1.0667	4.0	4048	0.9756
1.0345	5.0	5060	0.9478
1.0112	6.0	6072	0.9292
0.9922	7.0	7084	0.9137
0.9762	8.0	8096	0.9056
0.9627	9.0	9108	0.8977
0.9507	10.0	10120	0.8868
0.9411	11.0	11132	0.8823
0.9344	12.0	12144	0.8745
0.9261	13.0	13156	0.8688
0.9189	14.0	14168	0.8614
0.9133	15.0	15180	0.8609
0.9078	16.0	16192	0.8581
0.906	17.0	17204	0.8544
0.9015	18.0	18216	0.8537
0.8988	19.0	19228	0.8494
0.8975	20.0	20240	0.8491

Framework versions

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