N_roberta_imdb_padding20model

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

• Loss: 0.5709
• Accuracy: 0.9526

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
0.2052	1.0	1563	0.1966	0.9395
0.1578	2.0	3126	0.1547	0.9501
0.1132	3.0	4689	0.2315	0.9490
0.0801	4.0	6252	0.2392	0.9478
0.0455	5.0	7815	0.3256	0.9475
0.0377	6.0	9378	0.3895	0.9394
0.0299	7.0	10941	0.3465	0.9486
0.0199	8.0	12504	0.3895	0.9427
0.0232	9.0	14067	0.3813	0.945
0.0158	10.0	15630	0.4284	0.9476
0.0122	11.0	17193	0.4631	0.943
0.0094	12.0	18756	0.4639	0.9500
0.0074	13.0	20319	0.4256	0.9509
0.0032	14.0	21882	0.4599	0.9520
0.002	15.0	23445	0.5557	0.949
0.0025	16.0	25008	0.5381	0.9490
0.0018	17.0	26571	0.5017	0.9514
0.0008	18.0	28134	0.5676	0.9506
0.0	19.0	29697	0.5757	0.9519
0.0018	20.0	31260	0.5709	0.9526

Framework versions

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