Image-Arousal-new

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6535
• Accuracy: 0.4591

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: 0.0002
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2322	0.1855	100	1.2411	0.4452
1.1613	0.3711	200	1.2600	0.3987
1.2851	0.5566	300	1.2428	0.4052
1.1931	0.7421	400	1.2041	0.4559
1.1098	0.9276	500	1.1918	0.4586
1.1714	1.1132	600	1.1806	0.4721
1.1216	1.2987	700	1.1692	0.4651
1.2208	1.4842	800	1.1801	0.4614
1.0644	1.6698	900	1.1775	0.4596
1.1638	1.8553	1000	1.2031	0.4721
0.9559	2.0408	1100	1.2392	0.4521
0.8442	2.2263	1200	1.2544	0.4661
0.8713	2.4119	1300	1.2792	0.4744
0.8442	2.5974	1400	1.2618	0.4647
0.831	2.7829	1500	1.3202	0.4554
0.7774	2.9685	1600	1.3087	0.4572
0.5501	3.1540	1700	1.4975	0.4600
0.6069	3.3395	1800	1.5869	0.4512
0.4397	3.5250	1900	1.6458	0.4387
0.4468	3.7106	2000	1.6341	0.4493
0.4198	3.8961	2100	1.6535	0.4591

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1