N_bert_imdb_padding20model

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: 0.6296
• 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
0.2224	1.0	1563	0.2481	0.9156
0.1564	2.0	3126	0.2097	0.9386
0.0937	3.0	4689	0.2687	0.9347
0.056	4.0	6252	0.2934	0.9377
0.0382	5.0	7815	0.3589	0.9370
0.0304	6.0	9378	0.4149	0.9355
0.02	7.0	10941	0.4875	0.9348
0.0174	8.0	12504	0.4710	0.938
0.0135	9.0	14067	0.5213	0.9333
0.0122	10.0	15630	0.5069	0.9366
0.0132	11.0	17193	0.5394	0.9326
0.0067	12.0	18756	0.5691	0.9377
0.0023	13.0	20319	0.5857	0.9368
0.0043	14.0	21882	0.5734	0.9395
0.0048	15.0	23445	0.5936	0.9388
0.0046	16.0	25008	0.5803	0.9389
0.0032	17.0	26571	0.5693	0.9402
0.0006	18.0	28134	0.6308	0.9396
0.0016	19.0	29697	0.6221	0.9402
0.0008	20.0	31260	0.6296	0.9405

Framework versions

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