resnet-50-resnet50_fashion

This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1064
• Accuracy: 0.9740

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 7
• total_train_batch_size: 56
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6532	0.99	86	0.6781	0.6345
0.5407	2.0	173	0.5222	0.8590
0.4086	2.99	259	0.3595	0.8924
0.3449	4.0	346	0.2616	0.9184
0.3518	4.99	432	0.2288	0.9443
0.308	6.0	519	0.2758	0.9425
0.3209	7.0	606	0.3777	0.9369
0.284	7.99	692	0.1704	0.9555
0.2466	9.0	779	0.1571	0.9462
0.3123	9.99	865	0.6492	0.9406
0.2827	11.0	952	0.4968	0.9406
0.2736	11.99	1038	0.1370	0.9592
0.2476	13.0	1125	0.1616	0.9499
0.195	14.0	1212	0.1362	0.9610
0.2536	14.99	1298	0.1298	0.9536
0.2022	16.0	1385	0.7470	0.9518
0.2406	16.99	1471	0.1241	0.9647
0.2019	18.0	1558	0.1278	0.9536
0.2073	18.99	1644	0.1134	0.9685
0.1873	20.0	1731	0.6738	0.9629
0.2446	21.0	1818	0.1033	0.9685
0.1999	21.99	1904	0.1181	0.9647
0.1716	23.0	1991	0.1099	0.9610
0.175	23.99	2077	0.1064	0.9740
0.1962	25.0	2164	0.1174	0.9722
0.1943	25.99	2250	1.0625	0.9518
0.2044	27.0	2337	0.8419	0.9573
0.1835	28.0	2424	0.1112	0.9703
0.191	28.99	2510	0.1142	0.9685
0.1676	30.0	2597	0.1080	0.9647
0.1533	30.99	2683	0.1494	0.9647
0.1991	32.0	2770	0.1000	0.9703
0.1845	32.99	2856	0.0989	0.9740
0.1605	34.0	2943	0.0975	0.9685
0.1928	35.0	3030	0.4555	0.9629
0.1506	35.99	3116	0.1059	0.9703
0.1912	37.0	3203	0.1016	0.9647
0.1689	37.99	3289	0.5421	0.9666
0.1467	39.0	3376	0.1095	0.9647
0.1513	39.99	3462	0.3828	0.9703
0.1768	41.0	3549	0.0945	0.9703
0.1633	42.0	3636	0.2250	0.9592
0.1945	42.99	3722	0.2015	0.9685
0.1896	44.0	3809	0.1114	0.9666
0.1629	44.99	3895	0.0954	0.9666
0.1825	46.0	3982	0.0974	0.9740
0.1664	46.99	4068	0.0939	0.9703
0.1535	48.0	4155	0.0935	0.9722
0.1801	49.0	4242	0.0999	0.9703
0.1502	49.67	4300	0.1959	0.9703

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3