vit-base-patch32-384-finetuned-humid-classes-30

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: 0.3627
• Accuracy: 0.9730
• F1 Macro: 0.9481
• Precision Macro: 0.9667
• Recall Macro: 0.9444
• 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: 1.0
• F1 Humid: 1.0
• Precision Lump: 1.0
• Recall Lump: 1.0
• F1 Lump: 1.0
• Precision Moist: 0.8
• Recall Moist: 1.0
• F1 Moist: 0.8889
• 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 Moist	Recall Moist	F1 Moist	Precision Rockies	Recall Rockies	F1 Rockies
No log	1.0	3	1.7161	0.3784	0.2893	0.3111	0.4118	0.0	0.0	0.0	1.0	0.3636	0.5333	0.3333	0.8571	0.48	0.0	0.0	0.0	0.2	0.25	0.2222	0.3333	1.0	0.5
No log	2.0	6	1.3962	0.6216	0.4520	0.3958	0.5833	0.0	0.0	0.0	1.0	1.0	1.0	0.5	1.0	0.6667	0.0	0.0	0.0	0.5	0.5	0.5	0.375	1.0	0.5455
No log	3.0	9	1.0720	0.7027	0.6015	0.7519	0.6667	1.0	0.1667	0.2857	1.0	1.0	1.0	0.6364	1.0	0.7778	1.0	0.3333	0.5	0.5	0.5	0.5	0.375	1.0	0.5455
1.5274	4.0	12	0.7791	0.8108	0.7690	0.8333	0.7778	1.0	0.8333	0.9091	1.0	1.0	1.0	0.5833	1.0	0.7368	0.6667	0.3333	0.4444	1.0	0.5	0.6667	0.75	1.0	0.8571
1.5274	5.0	15	0.4969	0.9189	0.8928	0.9375	0.8750	0.75	1.0	0.8571	1.0	1.0	1.0	0.875	1.0	0.9333	1.0	0.8333	0.9091	1.0	0.75	0.8571	1.0	0.6667	0.8
1.5274	6.0	18	0.3627	0.9730	0.9481	0.9667	0.9444	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.8	1.0	0.8889	1.0	0.6667	0.8
0.5728	7.0	21	0.1932	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.5728	8.0	24	0.2180	0.9730	0.9481	0.9667	0.9444	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.8	1.0	0.8889	1.0	0.6667	0.8
0.5728	9.0	27	0.1210	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.1301	10.0	30	0.1074	0.9730	0.9481	0.9667	0.9444	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.8	1.0	0.8889	1.0	0.6667	0.8
0.1301	11.0	33	0.0750	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.1301	12.0	36	0.1066	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.1301	13.0	39	0.1137	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.021	14.0	42	0.1193	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.021	15.0	45	0.1274	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.021	16.0	48	0.1420	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0045	17.0	51	0.1530	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0045	18.0	54	0.1629	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0045	19.0	57	0.1720	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0023	20.0	60	0.1747	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0023	21.0	63	0.1689	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0023	22.0	66	0.1575	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0023	23.0	69	0.1469	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0016	24.0	72	0.1376	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0016	25.0	75	0.1327	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0016	26.0	78	0.1322	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0012	27.0	81	0.1362	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0012	28.0	84	0.1398	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0012	29.0	87	0.1431	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0011	30.0	90	0.1442	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0011	31.0	93	0.1462	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0011	32.0	96	0.1474	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0011	33.0	99	0.1475	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.001	34.0	102	0.1471	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.001	35.0	105	0.1468	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.001	36.0	108	0.1475	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	37.0	111	0.1487	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	38.0	114	0.1502	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	39.0	117	0.1514	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	40.0	120	0.1523	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	41.0	123	0.1530	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	42.0	126	0.1533	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0009	43.0	129	0.1531	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	44.0	132	0.1519	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	45.0	135	0.1511	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	46.0	138	0.1506	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	47.0	141	0.1504	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	48.0	144	0.1505	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	49.0	147	0.1504	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.6667	0.8
0.0008	50.0	150	0.1504	0.9730	0.9538	0.9762	0.9444	0.8571	1.0	0.9231	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	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