aift-model

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.1789
• Accuracy Thresh: 0.9161

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Thresh
No log	1.0	129	2.3568	0.3224
No log	2.0	258	1.6533	0.5518
No log	3.0	387	1.3939	0.7174
2.7357	4.0	516	1.3937	0.7875
2.7357	5.0	645	1.3416	0.8023
2.7357	6.0	774	1.3744	0.8696
2.7357	7.0	903	1.4840	0.8753
0.727	8.0	1032	1.6799	0.8858
0.727	9.0	1161	1.6802	0.8830
0.727	10.0	1290	1.8544	0.8968
0.727	11.0	1419	1.8931	0.8971
0.4047	12.0	1548	2.1624	0.8978
0.4047	13.0	1677	2.3448	0.9042
0.4047	14.0	1806	2.4661	0.9087
0.4047	15.0	1935	2.5098	0.9087
0.2396	16.0	2064	2.7352	0.9130
0.2396	17.0	2193	2.7756	0.9123
0.2396	18.0	2322	2.9367	0.9151
0.2396	19.0	2451	2.9757	0.9140
0.1658	20.0	2580	2.9550	0.9140
0.1658	21.0	2709	3.0914	0.9161
0.1658	22.0	2838	3.1505	0.9158
0.1658	23.0	2967	3.0856	0.9147
0.1233	24.0	3096	3.1968	0.9175
0.1233	25.0	3225	3.1789	0.9161

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2