N_distilbert_imdb_padding80model

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: 0.7459
• Accuracy: 0.9296

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
0.2421	1.0	1563	0.2031	0.9236
0.1775	2.0	3126	0.2252	0.9267
0.1171	3.0	4689	0.2886	0.9266
0.0694	4.0	6252	0.3525	0.9276
0.0479	5.0	7815	0.4599	0.9197
0.0401	6.0	9378	0.5028	0.9204
0.0333	7.0	10941	0.5221	0.9242
0.028	8.0	12504	0.5112	0.9247
0.0219	9.0	14067	0.5318	0.9273
0.0176	10.0	15630	0.6407	0.9206
0.0145	11.0	17193	0.5774	0.9280
0.0107	12.0	18756	0.6352	0.9236
0.0087	13.0	20319	0.6393	0.9293
0.0093	14.0	21882	0.6113	0.9297
0.0039	15.0	23445	0.6722	0.9304
0.0045	16.0	25008	0.6754	0.9295
0.0001	17.0	26571	0.7190	0.9295
0.0003	18.0	28134	0.7310	0.9301
0.0011	19.0	29697	0.7402	0.9297
0.0017	20.0	31260	0.7459	0.9296

Framework versions

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