N_distilbert_imdb_padding50model

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

• Loss: 0.7391
• Accuracy: 0.9327

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
0.2431	1.0	1563	0.2690	0.9089
0.1734	2.0	3126	0.2418	0.9260
0.1167	3.0	4689	0.4345	0.9078
0.0685	4.0	6252	0.3717	0.926
0.0445	5.0	7815	0.4502	0.9242
0.0338	6.0	9378	0.4786	0.9287
0.0293	7.0	10941	0.5332	0.9214
0.0191	8.0	12504	0.5435	0.9287
0.0182	9.0	14067	0.5450	0.9265
0.015	10.0	15630	0.5398	0.9297
0.0122	11.0	17193	0.6565	0.9226
0.0089	12.0	18756	0.6521	0.9280
0.0081	13.0	20319	0.6755	0.9285
0.0067	14.0	21882	0.6753	0.93
0.0054	15.0	23445	0.7014	0.9305
0.0023	16.0	25008	0.7440	0.9308
0.0004	17.0	26571	0.7371	0.9286
0.0	18.0	28134	0.7497	0.9302
0.0004	19.0	29697	0.7386	0.9324
0.0002	20.0	31260	0.7391	0.9327

Framework versions

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