roberta_sst2_padding0model

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

• Loss: 0.4539
• Accuracy: 0.9484

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
No log	1.0	433	0.1891	0.9407
0.3324	2.0	866	0.3948	0.9176
0.1922	3.0	1299	0.2418	0.9379
0.126	4.0	1732	0.3080	0.9407
0.069	5.0	2165	0.4075	0.9396
0.0358	6.0	2598	0.3955	0.9418
0.0298	7.0	3031	0.4060	0.9429
0.0298	8.0	3464	0.4284	0.9379
0.0207	9.0	3897	0.4804	0.9401
0.0197	10.0	4330	0.5089	0.9347
0.0177	11.0	4763	0.5430	0.9336
0.0143	12.0	5196	0.4997	0.9385
0.0138	13.0	5629	0.4695	0.9429
0.0066	14.0	6062	0.5391	0.9363
0.0066	15.0	6495	0.5354	0.9412
0.0042	16.0	6928	0.4295	0.9473
0.0067	17.0	7361	0.4948	0.9429
0.0053	18.0	7794	0.4720	0.9473
0.0041	19.0	8227	0.4552	0.9451
0.0068	20.0	8660	0.4539	0.9484

Framework versions

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