N_bert_imdb_padding30model

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: 0.6907
• Accuracy: 0.9394

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
0.2178	1.0	1563	0.2389	0.9234
0.1617	2.0	3126	0.2474	0.9303
0.0872	3.0	4689	0.3029	0.9283
0.065	4.0	6252	0.3493	0.9316
0.0348	5.0	7815	0.3685	0.9365
0.0311	6.0	9378	0.4913	0.9310
0.0205	7.0	10941	0.4485	0.9362
0.0177	8.0	12504	0.4903	0.9354
0.0147	9.0	14067	0.5786	0.9322
0.0119	10.0	15630	0.5245	0.9356
0.01	11.0	17193	0.5730	0.9364
0.0091	12.0	18756	0.5730	0.9383
0.006	13.0	20319	0.5596	0.9386
0.004	14.0	21882	0.6760	0.9354
0.0018	15.0	23445	0.5813	0.9402
0.0018	16.0	25008	0.6526	0.9378
0.0035	17.0	26571	0.6453	0.9384
0.0002	18.0	28134	0.6714	0.9392
0.0001	19.0	29697	0.6893	0.9397
0.0	20.0	31260	0.6907	0.9394

Framework versions

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