distilbert_distilbert-base-uncased-15-epoch

This model is a fine-tuned version of distilbert/distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 6.5739
• Accuracy: 0.4129

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2998	1.0	5714	1.3409	0.3419
1.1366	2.0	11428	1.3753	0.3634
0.9035	3.0	17142	1.4999	0.3794
0.6883	4.0	22856	1.7468	0.3906
0.5177	5.0	28570	2.1011	0.3997
0.3968	6.0	34284	2.2910	0.4009
0.2936	7.0	39998	2.8000	0.4121
0.2403	8.0	45712	2.7002	0.4086
0.1843	9.0	51426	3.4014	0.4160
0.1478	10.0	57140	3.5690	0.4057
0.1231	11.0	62854	4.3858	0.4102
0.1027	12.0	68568	4.6814	0.4124
0.0897	13.0	74282	4.9962	0.4186
0.0823	14.0	79996	6.0199	0.4160
0.0631	15.0	85710	6.5739	0.4129

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.0