N_bert_imdb_padding0model

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

• Loss: 0.6575
• Accuracy: 0.9405

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
0.2204	1.0	1563	0.2086	0.9332
0.1501	2.0	3126	0.2195	0.9356
0.0871	3.0	4689	0.3156	0.935
0.0555	4.0	6252	0.3170	0.9314
0.0362	5.0	7815	0.3568	0.9353
0.0282	6.0	9378	0.4438	0.9380
0.0199	7.0	10941	0.4900	0.9357
0.0219	8.0	12504	0.4963	0.9344
0.0115	9.0	14067	0.5554	0.9333
0.0078	10.0	15630	0.5974	0.9340
0.0087	11.0	17193	0.6081	0.9360
0.0038	12.0	18756	0.5909	0.9322
0.0096	13.0	20319	0.6002	0.9381
0.0061	14.0	21882	0.5645	0.9372
0.0057	15.0	23445	0.6415	0.9388
0.0019	16.0	25008	0.6901	0.9388
0.0005	17.0	26571	0.7099	0.9389
0.0	18.0	28134	0.7022	0.9392
0.0008	19.0	29697	0.6640	0.9398
0.0	20.0	31260	0.6575	0.9405

Framework versions

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