N_roberta_imdb_padding40model

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

• Loss: 0.4922
• Accuracy: 0.9495

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
0.2081	1.0	1563	0.2432	0.9283
0.1726	2.0	3126	0.1724	0.9493
0.114	3.0	4689	0.2842	0.9384
0.0767	4.0	6252	0.2583	0.9463
0.0552	5.0	7815	0.3703	0.9420
0.0357	6.0	9378	0.3342	0.9386
0.0318	7.0	10941	0.3284	0.9462
0.0316	8.0	12504	0.4194	0.9410
0.0149	9.0	14067	0.4083	0.9483
0.0175	10.0	15630	0.4237	0.9468
0.0151	11.0	17193	0.4459	0.9457
0.0113	12.0	18756	0.4569	0.9478
0.0061	13.0	20319	0.4325	0.9482
0.0034	14.0	21882	0.5188	0.9472
0.0059	15.0	23445	0.4740	0.9484
0.0078	16.0	25008	0.4421	0.9485
0.0	17.0	26571	0.4819	0.9493
0.0035	18.0	28134	0.4845	0.9492
0.0	19.0	29697	0.5065	0.9486
0.0013	20.0	31260	0.4922	0.9495

Framework versions

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