resnet-18-v3

This model is a fine-tuned version of microsoft/resnet-18 on an unknown dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 128
eval_batch_size: 128
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 512
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 176
num_epochs: 8

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Tp	Tn	Fp	Fn
6.4685	0.7928	22	1.1777	0.7029	0.6201	0.9206	0.7410	1508	986	924	130
3.9094	1.5766	44	0.7011	0.8602	0.8183	0.8962	0.8555	1468	1584	326	170
2.5221	2.3604	66	0.5245	0.9160	0.9323	0.8822	0.9065	1445	1805	105	193
2.0654	3.1441	88	0.4849	0.9180	0.9016	0.9231	0.9122	1512	1745	165	126
1.8586	3.9369	110	0.4350	0.9343	0.9281	0.9298	0.9289	1523	1792	118	115
1.7348	4.7207	132	0.4187	0.9270	0.8997	0.9475	0.9230	1552	1737	173	86
1.5149	5.5045	154	0.3705	0.9340	0.9086	0.9530	0.9303	1561	1753	157	77
1.5157	6.2883	176	0.2792	0.9614	0.9618	0.9542	0.9580	1563	1848	62	75
1.5215	7.0721	198	0.2364	0.9682	0.9757	0.9548	0.9651	1564	1871	39	74
1.4679	7.8649	220	0.2625	0.9625	0.9631	0.9554	0.9592	1565	1850	60	73