car_manufacturer_model

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

• Loss: 2.7826
• Accuracy: 0.3394

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	7	3.1387	0.2018
2.8998	2.0	14	3.1029	0.2018
2.7326	3.0	21	3.0453	0.2294
2.7326	4.0	28	3.0104	0.2385
2.5797	5.0	35	2.9655	0.2477
2.4873	6.0	42	2.9166	0.3211
2.4873	7.0	49	2.9122	0.2569
2.3408	8.0	56	2.8122	0.3119
2.2696	9.0	63	2.8159	0.3578
2.1527	10.0	70	2.8589	0.2752
2.1527	11.0	77	2.8248	0.2936
2.0649	12.0	84	2.7709	0.2936
2.0855	13.0	91	2.8183	0.2477
2.0855	14.0	98	2.7552	0.2569
1.9347	15.0	105	2.7826	0.3394

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0