N_roberta_imdb_padding0model

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

• Loss: 0.5120
• Accuracy: 0.9528

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
0.2059	1.0	1563	0.1926	0.9406
0.1663	2.0	3126	0.1661	0.9497
0.1024	3.0	4689	0.2261	0.9503
0.0691	4.0	6252	0.2616	0.9494
0.043	5.0	7815	0.3541	0.9466
0.0431	6.0	9378	0.3030	0.9522
0.0247	7.0	10941	0.3501	0.9482
0.0299	8.0	12504	0.3723	0.9478
0.02	9.0	14067	0.3530	0.9496
0.0204	10.0	15630	0.3836	0.9458
0.0144	11.0	17193	0.3471	0.9505
0.0095	12.0	18756	0.3696	0.9504
0.0087	13.0	20319	0.3877	0.9487
0.0078	14.0	21882	0.4487	0.9504
0.0056	15.0	23445	0.4657	0.9512
0.0032	16.0	25008	0.5004	0.9502
0.0041	17.0	26571	0.4863	0.9525
0.0036	18.0	28134	0.4833	0.9516
0.0	19.0	29697	0.5014	0.9523
0.0001	20.0	31260	0.5120	0.9528

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3