N_roberta_imdb_padding80model

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

• Loss: 0.5564
• Accuracy: 0.9511

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
0.2029	1.0	1563	0.2612	0.9242
0.1698	2.0	3126	0.1708	0.9496
0.1184	3.0	4689	0.2432	0.945
0.0767	4.0	6252	0.2460	0.9470
0.0527	5.0	7815	0.3357	0.9428
0.0398	6.0	9378	0.3318	0.9466
0.0248	7.0	10941	0.3870	0.9424
0.0236	8.0	12504	0.3998	0.9472
0.0214	9.0	14067	0.4392	0.9463
0.0171	10.0	15630	0.3822	0.9468
0.0167	11.0	17193	0.4762	0.9478
0.0078	12.0	18756	0.4300	0.948
0.0084	13.0	20319	0.5198	0.9473
0.003	14.0	21882	0.5064	0.9473
0.0079	15.0	23445	0.4918	0.9489
0.0036	16.0	25008	0.5179	0.9496
0.0008	17.0	26571	0.5576	0.9502
0.002	18.0	28134	0.5636	0.9500
0.003	19.0	29697	0.5534	0.9511
0.0015	20.0	31260	0.5564	0.9511

Framework versions

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