N_distilbert_imdb_padding30model

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: 0.7513
• Accuracy: 0.9320

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
0.2412	1.0	1563	0.2749	0.9004
0.1694	2.0	3126	0.2355	0.9270
0.1055	3.0	4689	0.3029	0.9262
0.0621	4.0	6252	0.3240	0.9282
0.0422	5.0	7815	0.4462	0.9269
0.0366	6.0	9378	0.4963	0.9274
0.0309	7.0	10941	0.5017	0.9286
0.0189	8.0	12504	0.6588	0.9198
0.0217	9.0	14067	0.5946	0.9218
0.02	10.0	15630	0.6104	0.9248
0.0112	11.0	17193	0.5921	0.9293
0.0096	12.0	18756	0.6499	0.9290
0.0075	13.0	20319	0.6577	0.9299
0.0036	14.0	21882	0.6225	0.9289
0.0043	15.0	23445	0.6558	0.9290
0.0015	16.0	25008	0.6923	0.9314
0.0036	17.0	26571	0.7606	0.9284
0.0	18.0	28134	0.7696	0.931
0.0028	19.0	29697	0.7493	0.9319
0.0005	20.0	31260	0.7513	0.9320

Framework versions

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