distilbert_imdb_padding0model

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: 0.7541
• Accuracy: 0.9328

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
0.2321	1.0	1563	0.2211	0.9195
0.1748	2.0	3126	0.2320	0.9289
0.1084	3.0	4689	0.3254	0.9251
0.0715	4.0	6252	0.3303	0.9267
0.0433	5.0	7815	0.4353	0.9276
0.0335	6.0	9378	0.4458	0.9302
0.033	7.0	10941	0.4704	0.9282
0.0171	8.0	12504	0.5326	0.9281
0.0147	9.0	14067	0.5456	0.9292
0.0099	10.0	15630	0.6037	0.9274
0.0166	11.0	17193	0.5636	0.9286
0.0101	12.0	18756	0.6355	0.9276
0.0086	13.0	20319	0.6102	0.9288
0.0068	14.0	21882	0.6305	0.9331
0.005	15.0	23445	0.6391	0.9293
0.0009	16.0	25008	0.7000	0.9339
0.0035	17.0	26571	0.7205	0.9325
0.0017	18.0	28134	0.7649	0.9294
0.0007	19.0	29697	0.7745	0.9329
0.0023	20.0	31260	0.7541	0.9328

Framework versions

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