pvt-tiny-224

This model is a fine-tuned version of Zetatech/pvt-tiny-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.4869
• Accuracy: 0.7833
• Precision: 0.7681
• Recall: 0.7833
• F1 Score: 0.7632

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
No log	1.0	4	0.5984	0.7333	0.5378	0.7333	0.6205
No log	2.0	8	0.6103	0.7333	0.5378	0.7333	0.6205
No log	3.0	12	0.5861	0.7333	0.5378	0.7333	0.6205
No log	4.0	16	0.5478	0.7333	0.5378	0.7333	0.6205
No log	5.0	20	0.5961	0.725	0.7119	0.725	0.7171
No log	6.0	24	0.5317	0.7542	0.7261	0.7542	0.7159
No log	7.0	28	0.5620	0.7458	0.7289	0.7458	0.7342
0.5878	8.0	32	0.5281	0.7542	0.7316	0.7542	0.6973
0.5878	9.0	36	0.5434	0.7625	0.7395	0.7625	0.7368
0.5878	10.0	40	0.5236	0.775	0.7658	0.775	0.7321
0.5878	11.0	44	0.5411	0.7542	0.7382	0.7542	0.7429
0.5878	12.0	48	0.5186	0.7708	0.7507	0.7708	0.7460
0.5878	13.0	52	0.5194	0.7667	0.7500	0.7667	0.7533
0.5878	14.0	56	0.5049	0.7875	0.7739	0.7875	0.7621
0.4973	15.0	60	0.5125	0.7833	0.7691	0.7833	0.7709
0.4973	16.0	64	0.5000	0.7917	0.7804	0.7917	0.7656
0.4973	17.0	68	0.5137	0.7583	0.7560	0.7583	0.7571
0.4973	18.0	72	0.4833	0.8	0.788	0.8	0.7833
0.4973	19.0	76	0.4929	0.7917	0.7816	0.7917	0.7843
0.4973	20.0	80	0.4858	0.8042	0.7930	0.8042	0.7887
0.4973	21.0	84	0.4900	0.7917	0.7777	0.7917	0.7743
0.4973	22.0	88	0.4886	0.7958	0.7829	0.7958	0.7815
0.439	23.0	92	0.4841	0.7917	0.7778	0.7917	0.7723
0.439	24.0	96	0.4855	0.8	0.7883	0.8	0.7885
0.439	25.0	100	0.4856	0.8	0.7879	0.8	0.7869
0.439	26.0	104	0.4839	0.8	0.7879	0.8	0.7869
0.439	27.0	108	0.4811	0.8	0.7879	0.8	0.7869
0.439	28.0	112	0.4834	0.8	0.7889	0.8	0.7901
0.439	29.0	116	0.4839	0.8	0.7889	0.8	0.7901
0.4092	30.0	120	0.4838	0.8	0.7889	0.8	0.7901

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3