distilbert_imdb_padding70model

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: 0.7503
• Accuracy: 0.9304

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
0.2385	1.0	1563	0.2404	0.9121
0.1808	2.0	3126	0.2250	0.9273
0.1148	3.0	4689	0.3192	0.9253
0.0711	4.0	6252	0.3479	0.9272
0.0566	5.0	7815	0.4082	0.9260
0.0409	6.0	9378	0.4675	0.9268
0.028	7.0	10941	0.4345	0.9268
0.0209	8.0	12504	0.5870	0.9224
0.0228	9.0	14067	0.4920	0.9269
0.0132	10.0	15630	0.5973	0.9248
0.013	11.0	17193	0.5962	0.9282
0.0127	12.0	18756	0.6695	0.9245
0.0028	13.0	20319	0.7241	0.9269
0.0102	14.0	21882	0.5935	0.9298
0.0033	15.0	23445	0.6654	0.9287
0.0039	16.0	25008	0.6902	0.9299
0.0012	17.0	26571	0.7281	0.9281
0.0023	18.0	28134	0.7369	0.9306
0.0001	19.0	29697	0.7462	0.9307
0.0019	20.0	31260	0.7503	0.9304

Framework versions

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