vit-base-patch32-384-finetuned-humid-classes-21

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: 0.1975
• Accuracy: 0.9535
• F1 Macro: 0.9210
• Precision Macro: 0.9429
• Recall Macro: 0.9222
• 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: 0.7143
• Recall Humid: 1.0
• F1 Humid: 0.8333
• Precision Lump: 1.0
• Recall Lump: 0.9444
• F1 Lump: 0.9714
• Precision Rockies: 1.0
• Recall Rockies: 0.6667
• F1 Rockies: 0.8

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	3	1.5583	0.2791	0.1633	0.2307	0.2143	0.0	0.0	0.0	0.32	0.5714	0.4103	0.0	0.0	0.0	0.75	0.1667	0.2727	0.0833	0.3333	0.1333
No log	2.0	6	1.1569	0.6744	0.3132	0.2760	0.3635	0.0	0.0	0.0	0.7647	0.9286	0.8387	0.0	0.0	0.0	0.6154	0.8889	0.7273	0.0	0.0	0.0
No log	3.0	9	0.9035	0.7907	0.4710	0.5251	0.48	0.0	0.0	0.0	0.9333	1.0	0.9655	1.0	0.4	0.5714	0.6923	1.0	0.8182	0.0	0.0	0.0
1.4163	4.0	12	0.7237	0.8140	0.7181	0.8372	0.6867	1.0	0.6667	0.8	0.875	1.0	0.9333	0.4286	0.6	0.5	0.8824	0.8333	0.8571	1.0	0.3333	0.5
1.4163	5.0	15	0.5206	0.7907	0.7406	0.8094	0.7022	1.0	0.6667	0.8	0.8235	1.0	0.9032	0.4	0.4	0.4	0.8235	0.7778	0.8	1.0	0.6667	0.8
1.4163	6.0	18	0.3811	0.8372	0.7872	0.8515	0.7533	1.0	0.6667	0.8	0.875	1.0	0.9333	0.5	0.6	0.5455	0.8824	0.8333	0.8571	1.0	0.6667	0.8
0.5555	7.0	21	0.3063	0.9070	0.8446	0.9048	0.8133	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8571	1.0	0.9231	1.0	0.6667	0.8
0.5555	8.0	24	0.2974	0.8837	0.8595	0.8875	0.86	1.0	1.0	1.0	1.0	1.0	1.0	0.5	0.8	0.6154	0.9375	0.8333	0.8824	1.0	0.6667	0.8
0.5555	9.0	27	0.3494	0.9070	0.8446	0.9048	0.8133	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8571	1.0	0.9231	1.0	0.6667	0.8
0.1337	10.0	30	0.2224	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.1337	11.0	33	0.2546	0.9070	0.8552	0.9117	0.8444	1.0	0.6667	0.8	0.9333	1.0	0.9655	0.625	1.0	0.7692	1.0	0.8889	0.9412	1.0	0.6667	0.8
0.1337	12.0	36	0.3058	0.9070	0.8446	0.9048	0.8133	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8571	1.0	0.9231	1.0	0.6667	0.8
0.1337	13.0	39	0.1613	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0206	14.0	42	0.1975	0.9535	0.9210	0.9429	0.9222	1.0	1.0	1.0	1.0	1.0	1.0	0.7143	1.0	0.8333	1.0	0.9444	0.9714	1.0	0.6667	0.8
0.0206	15.0	45	0.1585	0.9535	0.9210	0.9429	0.9222	1.0	1.0	1.0	1.0	1.0	1.0	0.7143	1.0	0.8333	1.0	0.9444	0.9714	1.0	0.6667	0.8
0.0206	16.0	48	0.1771	0.9070	0.8446	0.9048	0.8133	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.4	0.5	0.8571	1.0	0.9231	1.0	0.6667	0.8
0.0055	17.0	51	0.1757	0.8837	0.8278	0.8700	0.8022	1.0	1.0	1.0	1.0	1.0	1.0	0.5	0.4	0.4444	0.85	0.9444	0.8947	1.0	0.6667	0.8
0.0055	18.0	54	0.1714	0.9535	0.9210	0.9429	0.9222	1.0	1.0	1.0	1.0	1.0	1.0	0.7143	1.0	0.8333	1.0	0.9444	0.9714	1.0	0.6667	0.8
0.0055	19.0	57	0.1776	0.9535	0.9210	0.9429	0.9222	1.0	1.0	1.0	1.0	1.0	1.0	0.7143	1.0	0.8333	1.0	0.9444	0.9714	1.0	0.6667	0.8
0.0023	20.0	60	0.1818	0.9535	0.9210	0.9429	0.9222	1.0	1.0	1.0	1.0	1.0	1.0	0.7143	1.0	0.8333	1.0	0.9444	0.9714	1.0	0.6667	0.8
0.0023	21.0	63	0.1891	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0023	22.0	66	0.1977	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0023	23.0	69	0.2056	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0014	24.0	72	0.2097	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0014	25.0	75	0.2107	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0014	26.0	78	0.2088	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0011	27.0	81	0.2056	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0011	28.0	84	0.2020	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0011	29.0	87	0.1984	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0009	30.0	90	0.1965	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0009	31.0	93	0.1960	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0009	32.0	96	0.1957	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0009	33.0	99	0.1957	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0008	34.0	102	0.1959	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0008	35.0	105	0.1966	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0008	36.0	108	0.1966	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	37.0	111	0.1962	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	38.0	114	0.1961	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	39.0	117	0.1962	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	40.0	120	0.1966	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	41.0	123	0.1972	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	42.0	126	0.1976	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	43.0	129	0.1978	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	44.0	132	0.1981	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	45.0	135	0.1982	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	46.0	138	0.1985	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	47.0	141	0.1988	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	48.0	144	0.1989	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0007	49.0	147	0.1989	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8
0.0006	50.0	150	0.1989	0.9070	0.8638	0.8989	0.8422	1.0	1.0	1.0	1.0	1.0	1.0	0.6	0.6	0.6	0.8947	0.9444	0.9189	1.0	0.6667	0.8

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0