vit-base-patch32-384-finetuned-humid-classes-17

This model is a fine-tuned version of google/vit-base-patch32-384 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1444
• Accuracy: 0.9516
• F1 Macro: 0.9280
• Precision Macro: 0.9727
• Recall Macro: 0.9057
• Precision Dry: 1.0
• Recall Dry: 1.0
• F1 Dry: 1.0
• Precision Firm: 1.0
• Recall Firm: 1.0
• F1 Firm: 1.0
• Precision Humid: 1.0
• Recall Humid: 0.6
• F1 Humid: 0.75
• Precision Lump: 0.8636
• Recall Lump: 1.0
• F1 Lump: 0.9268
• Precision Rockies: 1.0
• Recall Rockies: 0.9286
• F1 Rockies: 0.9630

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• 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_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	Precision Macro	Recall Macro	Precision Dry	Recall Dry	F1 Dry	Precision Firm	Recall Firm	F1 Firm	Precision Humid	Recall Humid	F1 Humid	Precision Lump	Recall Lump	F1 Lump	Precision Rockies	Recall Rockies	F1 Rockies
No log	1.0	5	1.3844	0.4677	0.3495	0.4783	0.3809	0.6667	0.6	0.6316	1.0	0.2857	0.4444	0.0	0.0	0.0	0.3913	0.9474	0.5538	0.3333	0.0714	0.1176
1.3541	2.0	10	1.1354	0.6129	0.4513	0.4503	0.5095	1.0	0.6	0.75	0.7778	1.0	0.875	0.0	0.0	0.0	0.4737	0.9474	0.6316	0.0	0.0	0.0
1.3541	3.0	15	0.6608	0.8710	0.7994	0.9307	0.7866	1.0	1.0	1.0	0.9333	1.0	0.9655	1.0	0.2	0.3333	0.72	0.9474	0.8182	1.0	0.7857	0.88
0.6379	4.0	20	0.3818	0.9032	0.8630	0.952	0.8371	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.76	1.0	0.8636	1.0	0.7857	0.88
0.6379	5.0	25	0.3229	0.9032	0.8630	0.952	0.8371	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.76	1.0	0.8636	1.0	0.7857	0.88
0.1806	6.0	30	0.2452	0.9194	0.8756	0.9583	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.7917	1.0	0.8837	1.0	0.8571	0.9231
0.1806	7.0	35	0.1912	0.9194	0.9091	0.8974	0.9436	1.0	1.0	1.0	0.9333	1.0	0.9655	0.625	1.0	0.7692	1.0	0.7895	0.8824	0.9286	0.9286	0.9286
0.109	8.0	40	0.2545	0.9032	0.8630	0.952	0.8371	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.76	1.0	0.8636	1.0	0.7857	0.88
0.109	9.0	45	0.1644	0.9355	0.9293	0.9162	0.9579	1.0	1.0	1.0	0.9333	1.0	0.9655	0.7143	1.0	0.8333	1.0	0.7895	0.8824	0.9333	1.0	0.9655
0.046	10.0	50	0.1444	0.9516	0.9280	0.9727	0.9057	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.75	0.8636	1.0	0.9268	1.0	0.9286	0.9630
0.046	11.0	55	0.1762	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0106	12.0	60	0.1682	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0106	13.0	65	0.1277	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0045	14.0	70	0.2338	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0045	15.0	75	0.1420	0.9355	0.9147	0.9133	0.9209	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8	0.7273	0.9	0.9474	0.9231	1.0	0.8571	0.9231
0.0026	16.0	80	0.1907	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0026	17.0	85	0.2488	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0034	18.0	90	0.1712	0.9194	0.8939	0.8943	0.9066	1.0	1.0	1.0	1.0	1.0	1.0	0.5714	0.8	0.6667	0.9	0.9474	0.9231	1.0	0.7857	0.88
0.0034	19.0	95	0.1327	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.003	20.0	100	0.3557	0.9355	0.8878	0.9652	0.8657	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.8261	1.0	0.9048	1.0	0.9286	0.9630
0.003	21.0	105	0.2862	0.9355	0.8878	0.9652	0.8657	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.8261	1.0	0.9048	1.0	0.9286	0.9630
0.0015	22.0	110	0.0981	0.9516	0.9159	0.9310	0.9057	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.9048	1.0	0.95	1.0	0.9286	0.9630
0.0015	23.0	115	0.1134	0.9516	0.9353	0.925	0.9647	1.0	1.0	1.0	1.0	1.0	1.0	0.625	1.0	0.7692	1.0	0.8947	0.9444	1.0	0.9286	0.9630
0.0014	24.0	120	0.1602	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0014	25.0	125	0.3842	0.9194	0.8756	0.9583	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.7917	1.0	0.8837	1.0	0.8571	0.9231
0.0012	26.0	130	0.3105	0.9194	0.8756	0.9583	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.4	0.5714	0.7917	1.0	0.8837	1.0	0.8571	0.9231
0.0012	27.0	135	0.1442	0.9516	0.9280	0.9727	0.9057	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.75	0.8636	1.0	0.9268	1.0	0.9286	0.9630
0.0009	28.0	140	0.0838	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0009	29.0	145	0.0709	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0007	30.0	150	0.0723	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0007	31.0	155	0.0834	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0006	32.0	160	0.1004	0.9516	0.9346	0.9410	0.9314	1.0	1.0	1.0	1.0	1.0	1.0	0.8	0.8	0.8	0.9048	1.0	0.95	1.0	0.8571	0.9231
0.0006	33.0	165	0.1159	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0006	34.0	170	0.1305	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0006	35.0	175	0.1421	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0005	36.0	180	0.1511	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0005	37.0	185	0.1574	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0005	38.0	190	0.1620	0.9355	0.9033	0.9227	0.8914	1.0	1.0	1.0	1.0	1.0	1.0	0.75	0.6	0.6667	0.8636	1.0	0.9268	1.0	0.8571	0.9231
0.0005	39.0	195	0.1655	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	40.0	200	0.1684	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	41.0	205	0.1693	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	42.0	210	0.1701	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	43.0	215	0.1696	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	44.0	220	0.1688	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	45.0	225	0.1684	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0004	46.0	230	0.1686	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0004	47.0	235	0.1688	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0004	48.0	240	0.1688	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0004	49.0	245	0.1689	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231
0.0005	50.0	250	0.1688	0.9194	0.8656	0.8986	0.8514	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8261	1.0	0.9048	1.0	0.8571	0.9231

Framework versions

• Transformers 4.56.1
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0