deit_flyswot

This model was trained from scratch on the image_folder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0755
• F1: 0.9908

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: 32
• eval_batch_size: 32
• seed: 666
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	52	0.5710	0.8095
No log	2.0	104	0.2814	0.9380
No log	3.0	156	0.1719	0.9555
No log	4.0	208	0.1410	0.9692
No log	5.0	260	0.1457	0.9680
No log	6.0	312	0.1084	0.9747
No log	7.0	364	0.0892	0.9736
No log	8.0	416	0.0962	0.9831
No log	9.0	468	0.0819	0.9796
0.2034	10.0	520	0.0916	0.9778
0.2034	11.0	572	0.0793	0.9827
0.2034	12.0	624	0.0818	0.9894
0.2034	13.0	676	0.0852	0.9807
0.2034	14.0	728	0.0938	0.9778
0.2034	15.0	780	0.0814	0.9876
0.2034	16.0	832	0.0702	0.9892
0.2034	17.0	884	0.0801	0.9892
0.2034	18.0	936	0.0806	0.9892
0.2034	19.0	988	0.0769	0.9926
0.0115	20.0	1040	0.0800	0.9926
0.0115	21.0	1092	0.0794	0.9926
0.0115	22.0	1144	0.0762	0.9846
0.0115	23.0	1196	0.0789	0.9830
0.0115	24.0	1248	0.0794	0.9829
0.0115	25.0	1300	0.0770	0.9908
0.0115	26.0	1352	0.0791	0.9829
0.0115	27.0	1404	0.0813	0.9892
0.0115	28.0	1456	0.0816	0.9908
0.0058	29.0	1508	0.0774	0.9908
0.0058	30.0	1560	0.0755	0.9908

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.11.6