models

This model is a fine-tuned version of microsoft/resnet-18 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4704
• Accuracy: 0.8182

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4144	0.99	20	0.9938	0.7
0.7896	1.98	40	0.7022	0.7152
0.6191	2.96	60	0.6079	0.7636
0.6114	4.0	81	0.5554	0.7939
0.5365	4.99	101	0.5233	0.8152
0.4989	5.98	121	0.4934	0.8303
0.5111	6.96	141	0.5181	0.8
0.476	8.0	162	0.4844	0.8182
0.4655	8.99	182	0.4870	0.8152
0.4335	9.98	202	0.4802	0.8242
0.44	10.96	222	0.4776	0.8182
0.3989	12.0	243	0.4804	0.8182
0.4007	12.99	263	0.4768	0.8242
0.3987	13.98	283	0.4610	0.8303
0.3922	14.96	303	0.4578	0.8212
0.3924	16.0	324	0.4804	0.8182
0.3995	16.99	344	0.4736	0.8121
0.3623	17.98	364	0.4715	0.8121
0.3621	18.96	384	0.4671	0.8212
0.3629	19.75	400	0.4704	0.8182

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.0
• Tokenizers 0.13.3