resnet_4090_downsample_normal_2class

This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder 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: 0.0001
train_batch_size: 24
eval_batch_size: 4
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 15

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.308	1.0	342	0.3033	0.8653	0.8540	0.8653	0.8249
0.1871	2.0	684	0.3260	0.8662	0.8718	0.8662	0.8687
0.1234	3.0	1026	0.3255	0.8810	0.8823	0.8810	0.8817
0.097	4.0	1368	0.3200	0.8983	0.8958	0.8983	0.8969
0.0774	5.0	1710	0.4122	0.8709	0.8868	0.8709	0.8770
0.065	6.0	2052	0.4306	0.8829	0.8875	0.8829	0.8849
0.0573	7.0	2394	0.4371	0.8803	0.8878	0.8803	0.8835
0.0576	8.0	2736	0.4595	0.8818	0.8850	0.8818	0.8833
0.0496	9.0	3078	0.5312	0.8620	0.8791	0.8620	0.8686
0.0426	10.0	3420	0.5089	0.8713	0.8828	0.8713	0.8760
0.0433	11.0	3762	0.5283	0.8683	0.8824	0.8683	0.8739
0.0324	12.0	4104	0.5527	0.8638	0.8815	0.8638	0.8706
0.0339	13.0	4446	0.5878	0.8595	0.8831	0.8595	0.8681
0.0363	14.0	4788	0.5898	0.8632	0.8835	0.8632	0.8708
0.0302	15.0	5130	0.5975	0.8611	0.8820	0.8611	0.8689