N_distilbert_imdb_padding10model

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: 0.7558
• Accuracy: 0.9335

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
0.2364	1.0	1563	0.2100	0.9252
0.1754	2.0	3126	0.2253	0.928
0.1006	3.0	4689	0.2885	0.9287
0.0627	4.0	6252	0.3760	0.9271
0.0436	5.0	7815	0.4199	0.9281
0.0353	6.0	9378	0.4874	0.9294
0.0176	7.0	10941	0.6403	0.919
0.0172	8.0	12504	0.5783	0.9264
0.0206	9.0	14067	0.5343	0.9298
0.0125	10.0	15630	0.6186	0.927
0.012	11.0	17193	0.5948	0.9309
0.0094	12.0	18756	0.6524	0.9293
0.0095	13.0	20319	0.6730	0.9262
0.0053	14.0	21882	0.6670	0.9316
0.0024	15.0	23445	0.6873	0.9322
0.0028	16.0	25008	0.6858	0.9328
0.0007	17.0	26571	0.7114	0.9326
0.0014	18.0	28134	0.7477	0.9331
0.0	19.0	29697	0.7567	0.9335
0.0	20.0	31260	0.7558	0.9335

Framework versions

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