cards-top_left_swin-tiny-patch4-window7-224-finetuned-dough

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.9991
• Accuracy: 0.5875

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5983	1.0	1240	1.3399	0.4436
1.5587	2.0	2481	1.3298	0.4366
1.4955	3.0	3721	1.1679	0.5129
1.4884	4.0	4962	1.1331	0.5299
1.4413	5.0	6202	1.1286	0.5287
1.4395	6.0	7443	1.1316	0.5226
1.4497	7.0	8683	1.2127	0.4844
1.3988	8.0	9924	1.1119	0.5301
1.4457	9.0	11164	1.0984	0.5389
1.4153	10.0	12405	1.1226	0.5269
1.3962	11.0	13645	1.0610	0.5573
1.3911	12.0	14886	1.0540	0.5595
1.3617	13.0	16126	1.0646	0.5530
1.3766	14.0	17367	1.0722	0.5532
1.3693	15.0	18607	1.0243	0.5721
1.3624	16.0	19848	1.0212	0.5763
1.3638	17.0	21088	1.0667	0.5580
1.4007	18.0	22329	1.0314	0.5730
1.3415	19.0	23569	1.0191	0.5755
1.3802	20.0	24810	1.0142	0.5770
1.3572	21.0	26050	1.0125	0.5771
1.2962	22.0	27291	1.0167	0.5763
1.2831	23.0	28531	1.0043	0.5829
1.3272	24.0	29772	0.9990	0.5858
1.3197	25.0	31012	1.0033	0.5830
1.3203	26.0	32253	1.0075	0.5818
1.3172	27.0	33493	1.0008	0.5852
1.3197	28.0	34734	1.0016	0.5847
1.2879	29.0	35974	1.0017	0.5867
1.2907	29.99	37200	0.9991	0.5875

Framework versions

• Transformers 4.37.2
• Pytorch 2.0.1+cu117
• Datasets 2.17.0
• Tokenizers 0.15.2