deberta-v3-xsmall-finetuned-DAGPap22

This model is a fine-tuned version of microsoft/deberta-v3-xsmall on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0798
• Accuracy: 0.9907
• F1: 0.9934

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: 4.5e-05
• train_batch_size: 12
• eval_batch_size: 12
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	402	0.1626	0.9477	0.9616
0.4003	2.0	804	0.0586	0.9794	0.9853
0.1075	3.0	1206	0.0342	0.9907	0.9933
0.0581	4.0	1608	0.1140	0.9776	0.9838
0.0245	5.0	2010	0.1409	0.9776	0.9842
0.0245	6.0	2412	0.0732	0.9832	0.9881
0.0167	7.0	2814	0.1996	0.9682	0.9778
0.0139	8.0	3216	0.1219	0.9850	0.9894
0.006	9.0	3618	0.0670	0.9907	0.9934
0.0067	10.0	4020	0.1036	0.9869	0.9907
0.0067	11.0	4422	0.1220	0.9776	0.9838
0.0041	12.0	4824	0.1768	0.9776	0.9839
0.0007	13.0	5226	0.0943	0.9888	0.9920
0.0	14.0	5628	0.0959	0.9907	0.9934
0.0054	15.0	6030	0.0915	0.9888	0.9921
0.0054	16.0	6432	0.1618	0.9794	0.9855
0.0019	17.0	6834	0.0794	0.9907	0.9934
0.0	18.0	7236	0.0799	0.9907	0.9934
0.0	19.0	7638	0.0797	0.9907	0.9934
0.0	20.0	8040	0.0798	0.9907	0.9934

Framework versions

• Transformers 4.18.0
• Pytorch 1.11.0
• Datasets 2.1.0
• Tokenizers 0.12.1