resnet-50-image-classification

This model is a fine-tuned version of microsoft/resnet-50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3093
• Accuracy: 0.9408

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: 101010
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2000
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	338	2.2768	0.5172
2.2806	2.0	676	2.0111	0.6903
1.8538	3.0	1014	1.2525	0.7467
1.8538	4.0	1352	0.6251	0.8578
0.8758	5.0	1690	0.3761	0.8967
0.4181	6.0	2028	0.2852	0.9144
0.4181	7.0	2366	0.2492	0.9244
0.2458	8.0	2704	0.2169	0.9364
0.1721	9.0	3042	0.2121	0.9358
0.1721	10.0	3380	0.2052	0.9403
0.1089	11.0	3718	0.2075	0.9414
0.0783	12.0	4056	0.2164	0.9367
0.0783	13.0	4394	0.2274	0.9381
0.0573	14.0	4732	0.2196	0.9433
0.0465	15.0	5070	0.2415	0.9381
0.0465	16.0	5408	0.2370	0.9433
0.0375	17.0	5746	0.2521	0.94
0.0288	18.0	6084	0.2533	0.9411
0.0288	19.0	6422	0.2608	0.9381
0.0253	20.0	6760	0.2602	0.9397
0.0207	21.0	7098	0.2712	0.94
0.0207	22.0	7436	0.2584	0.9408
0.0187	23.0	7774	0.2703	0.9419
0.012	24.0	8112	0.2772	0.9422
0.012	25.0	8450	0.2712	0.9419
0.0174	26.0	8788	0.2774	0.9422
0.0137	27.0	9126	0.2857	0.9414
0.0137	28.0	9464	0.2796	0.9428
0.0111	29.0	9802	0.3008	0.9367
0.0106	30.0	10140	0.2938	0.9369
0.0106	31.0	10478	0.2863	0.9406
0.0079	32.0	10816	0.2903	0.9425
0.0078	33.0	11154	0.2961	0.9419
0.0078	34.0	11492	0.2882	0.9417
0.0056	35.0	11830	0.2974	0.9406
0.0041	36.0	12168	0.2997	0.9419
0.0039	37.0	12506	0.3123	0.9367
0.0039	38.0	12844	0.3009	0.9408
0.0036	39.0	13182	0.3009	0.9422
0.0055	40.0	13520	0.3093	0.9408

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1