N_roberta_imdb_padding100model

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

• Loss: 0.4217
• Accuracy: 0.9496

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
0.2097	1.0	1563	0.1893	0.9421
0.1725	2.0	3126	0.1588	0.9470
0.1091	3.0	4689	0.2349	0.9444
0.0796	4.0	6252	0.2474	0.9435
0.0543	5.0	7815	0.2866	0.9453
0.0387	6.0	9378	0.3219	0.9452
0.0252	7.0	10941	0.3410	0.9443
0.0278	8.0	12504	0.3811	0.9447
0.0234	9.0	14067	0.3518	0.9459
0.0177	10.0	15630	0.3733	0.9437
0.0247	11.0	17193	0.3569	0.9481
0.0118	12.0	18756	0.3711	0.9441
0.0166	13.0	20319	0.3597	0.9470
0.0109	14.0	21882	0.4184	0.9498
0.0046	15.0	23445	0.4363	0.9438
0.0046	16.0	25008	0.4597	0.9467
0.0057	17.0	26571	0.4282	0.9481
0.0045	18.0	28134	0.4345	0.9482
0.0056	19.0	29697	0.4097	0.9493
0.0022	20.0	31260	0.4217	0.9496

Framework versions

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