vit-msn-small-finetuned-galaxy10-decals

This model is a fine-tuned version of facebook/vit-msn-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3987
• Accuracy: 0.8729

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9097	0.9978	112	0.7858	0.7295
0.8257	1.9955	224	0.5902	0.7940
0.6911	2.9933	336	0.5490	0.8178
0.6293	4.0	449	0.5251	0.8184
0.555	4.9978	561	0.5009	0.8341
0.5486	5.9955	673	0.5087	0.8259
0.5012	6.9933	785	0.4373	0.8566
0.4853	8.0	898	0.4146	0.8679
0.4049	8.9978	1010	0.3987	0.8729
0.4463	9.9777	1120	0.3896	0.8710

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1