distilbert_imdb_padding40model

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: 0.7661
• 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
0.237	1.0	1563	0.2751	0.8951
0.1699	2.0	3126	0.2169	0.9289
0.1143	3.0	4689	0.3155	0.9232
0.0728	4.0	6252	0.3584	0.9266
0.0396	5.0	7815	0.4750	0.9234
0.0359	6.0	9378	0.5132	0.9194
0.0206	7.0	10941	0.4754	0.9283
0.0252	8.0	12504	0.5639	0.9235
0.0168	9.0	14067	0.5000	0.9281
0.0153	10.0	15630	0.5634	0.9283
0.0105	11.0	17193	0.6007	0.9303
0.0093	12.0	18756	0.6327	0.9277
0.0051	13.0	20319	0.7021	0.9268
0.0035	14.0	21882	0.6893	0.9290
0.0061	15.0	23445	0.7048	0.9300
0.002	16.0	25008	0.7754	0.9294
0.0032	17.0	26571	0.7550	0.9297
0.0006	18.0	28134	0.7681	0.9304
0.0006	19.0	29697	0.7683	0.9303
0.0017	20.0	31260	0.7661	0.9304

Framework versions

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