N_bert_imdb_padding100model

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: 0.7094
• Accuracy: 0.9372

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
0.2271	1.0	1563	0.2627	0.9098
0.1657	2.0	3126	0.2289	0.9296
0.0928	3.0	4689	0.3241	0.9247
0.0656	4.0	6252	0.3616	0.9314
0.045	5.0	7815	0.3673	0.9310
0.034	6.0	9378	0.4597	0.9307
0.0202	7.0	10941	0.4430	0.9364
0.0188	8.0	12504	0.5484	0.9296
0.0144	9.0	14067	0.5043	0.9344
0.0115	10.0	15630	0.5540	0.9336
0.0094	11.0	17193	0.4962	0.9338
0.0066	12.0	18756	0.5736	0.9364
0.0068	13.0	20319	0.6779	0.9335
0.0038	14.0	21882	0.6431	0.9355
0.0021	15.0	23445	0.6151	0.9359
0.0009	16.0	25008	0.7113	0.9348
0.0	17.0	26571	0.7734	0.9336
0.003	18.0	28134	0.7006	0.9372
0.0008	19.0	29697	0.7140	0.9374
0.0	20.0	31260	0.7094	0.9372

Framework versions

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