roberta_sst5_padding40model

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

• Loss: 3.7057
• Accuracy: 0.5620

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
1.2915	1.0	534	1.3392	0.4199
1.0328	2.0	1068	1.0073	0.5679
0.8737	3.0	1602	1.0071	0.5855
0.7544	4.0	2136	1.1617	0.5552
0.6062	5.0	2670	1.2311	0.5588
0.4991	6.0	3204	1.4888	0.5516
0.4032	7.0	3738	1.5889	0.5498
0.3291	8.0	4272	1.6908	0.5597
0.2557	9.0	4806	1.9047	0.5593
0.2262	10.0	5340	2.1743	0.5502
0.1997	11.0	5874	2.4720	0.5534
0.1719	12.0	6408	2.7418	0.5633
0.1584	13.0	6942	2.9747	0.5665
0.1423	14.0	7476	3.2228	0.5543
0.1016	15.0	8010	3.4270	0.5475
0.0865	16.0	8544	3.4458	0.5611
0.0655	17.0	9078	3.6376	0.5552
0.0486	18.0	9612	3.6493	0.5579
0.0446	19.0	10146	3.7012	0.5611
0.0395	20.0	10680	3.7057	0.5620

Framework versions

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