roberta_sst5_padding60model

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.8019
• Accuracy: 0.5674

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: 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2655	1.0	534	1.1383	0.5104
0.9941	2.0	1068	1.0110	0.5588
0.8336	3.0	1602	1.0624	0.5661
0.6969	4.0	2136	1.2062	0.5520
0.5586	5.0	2670	1.2584	0.5643
0.447	6.0	3204	1.5017	0.5624
0.3437	7.0	3738	1.7553	0.5493
0.2636	8.0	4272	1.8688	0.5557
0.224	9.0	4806	2.1154	0.5638
0.2058	10.0	5340	2.5540	0.5462
0.1692	11.0	5874	2.8222	0.5462
0.1631	12.0	6408	2.8802	0.5588
0.1285	13.0	6942	3.1257	0.5597
0.1158	14.0	7476	3.2779	0.5683
0.0863	15.0	8010	3.5119	0.5561
0.0836	16.0	8544	3.4229	0.5633
0.0604	17.0	9078	3.6220	0.5615
0.0391	18.0	9612	3.7717	0.5566
0.0399	19.0	10146	3.7460	0.5647
0.0279	20.0	10680	3.8019	0.5674

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1
• Datasets 2.12.0
• Tokenizers 0.13.3