This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.2813
F1 Macro: 0.4280
F1 Micro: 0.5455
F1 Weighted: 0.4882
Precision Macro: 0.4004
Precision Micro: 0.5455
Precision Weighted: 0.4529
Recall Macro: 0.4762
Recall Micro: 0.5455
Recall Weighted: 0.5455
Accuracy: 0.5455

Model description

Using a batch size of 16

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: 1e-05
train_batch_size: 8
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH 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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	F1 Micro	F1 Weighted	Precision Macro	Precision Micro	Precision Weighted	Recall Macro	Recall Micro	Recall Weighted	Accuracy
1.9371	1.0	29	1.9372	0.0504	0.1212	0.0604	0.0334	0.1212	0.0403	0.1029	0.1212	0.1212	0.1212
1.9078	2.0	58	1.9066	0.0454	0.1818	0.0602	0.0272	0.1818	0.0361	0.1371	0.1818	0.1818	0.1818
1.9276	3.0	87	1.8808	0.0696	0.1818	0.0968	0.0492	0.1818	0.0682	0.1295	0.1818	0.1818	0.1818
1.8373	4.0	116	1.8696	0.0485	0.2045	0.0695	0.0292	0.2045	0.0418	0.1429	0.2045	0.2045	0.2045
1.8152	5.0	145	1.8490	0.1339	0.2576	0.1745	0.1298	0.2576	0.1640	0.1944	0.2576	0.2576	0.2576
1.8488	6.0	174	1.8281	0.1379	0.2727	0.1817	0.1512	0.2727	0.1891	0.1997	0.2727	0.2727	0.2727
1.7626	7.0	203	1.7917	0.2271	0.3333	0.2718	0.1922	0.3333	0.2298	0.2783	0.3333	0.3333	0.3333
1.7169	8.0	232	1.7478	0.2887	0.4242	0.3465	0.2706	0.4242	0.3154	0.3426	0.4242	0.4242	0.4242
1.5364	9.0	261	1.7098	0.2835	0.4091	0.3409	0.2720	0.4091	0.3245	0.3324	0.4091	0.4091	0.4091
1.7373	10.0	290	1.6765	0.2906	0.4167	0.3463	0.2726	0.4167	0.3157	0.3386	0.4167	0.4167	0.4167
1.5345	11.0	319	1.6423	0.2805	0.3939	0.3342	0.3728	0.3939	0.4258	0.3275	0.3939	0.3939	0.3939
1.6421	12.0	348	1.6103	0.3324	0.4697	0.3978	0.4583	0.4697	0.5178	0.3760	0.4697	0.4697	0.4697
1.5266	13.0	377	1.5835	0.3171	0.4621	0.3822	0.2917	0.4621	0.3483	0.3748	0.4621	0.4621	0.4621
1.5182	14.0	406	1.5633	0.3133	0.4242	0.3680	0.3634	0.4242	0.4009	0.3568	0.4242	0.4242	0.4242
1.5341	15.0	435	1.5528	0.3015	0.4167	0.3585	0.3109	0.4167	0.3638	0.3499	0.4167	0.4167	0.4167
1.3961	16.0	464	1.5273	0.3449	0.4545	0.3991	0.4329	0.4545	0.4704	0.3839	0.4545	0.4545	0.4545
1.3601	17.0	493	1.4971	0.3670	0.5	0.4357	0.5047	0.5	0.5382	0.4078	0.5	0.5	0.5
1.2535	18.0	522	1.5006	0.3511	0.4621	0.4138	0.4778	0.4621	0.5101	0.3872	0.4621	0.4621	0.4621
1.2375	19.0	551	1.4659	0.3655	0.4924	0.4345	0.4298	0.4924	0.4797	0.4020	0.4924	0.4924	0.4924
1.2141	20.0	580	1.4407	0.3914	0.5076	0.4565	0.4650	0.5076	0.5087	0.4217	0.5076	0.5076	0.5076
1.2831	21.0	609	1.4454	0.3965	0.5152	0.4645	0.4801	0.5152	0.5265	0.4214	0.5152	0.5152	0.5152
1.1543	22.0	638	1.4167	0.4285	0.5455	0.4997	0.4781	0.5455	0.5309	0.4521	0.5455	0.5455	0.5455
1.4079	23.0	667	1.4465	0.3675	0.4621	0.4269	0.4187	0.4621	0.4676	0.3929	0.4621	0.4621	0.4621
1.0619	24.0	696	1.4249	0.4092	0.5076	0.4724	0.4659	0.5076	0.5180	0.4336	0.5076	0.5076	0.5076
1.1059	25.0	725	1.3834	0.4356	0.5530	0.5061	0.5025	0.5530	0.5491	0.4594	0.5530	0.5530	0.5530
1.192	26.0	754	1.3784	0.4286	0.5379	0.4893	0.4566	0.5379	0.4969	0.4544	0.5379	0.5379	0.5379
1.21	27.0	783	1.3874	0.4409	0.5379	0.5060	0.4709	0.5379	0.5258	0.4616	0.5379	0.5379	0.5379
1.0901	28.0	812	1.3621	0.4402	0.5379	0.5074	0.4635	0.5379	0.5204	0.4557	0.5379	0.5379	0.5379
1.1254	29.0	841	1.3714	0.4265	0.5227	0.4873	0.4492	0.5227	0.4984	0.4449	0.5227	0.5227	0.5227
0.9345	30.0	870	1.3525	0.4425	0.5379	0.5074	0.4736	0.5379	0.5264	0.4557	0.5379	0.5379	0.5379
1.2036	31.0	899	1.3592	0.4363	0.5379	0.5020	0.4869	0.5379	0.5368	0.4533	0.5379	0.5379	0.5379
1.036	32.0	928	1.3362	0.4451	0.5455	0.5109	0.4673	0.5455	0.5226	0.4637	0.5455	0.5455	0.5455
0.9979	33.0	957	1.3492	0.4454	0.5455	0.5134	0.4808	0.5455	0.5358	0.4620	0.5455	0.5455	0.5455
0.8353	34.0	986	1.3402	0.4635	0.5606	0.5301	0.4659	0.5606	0.5268	0.4854	0.5606	0.5606	0.5606
0.9384	35.0	1015	1.3414	0.4408	0.5455	0.5088	0.4664	0.5455	0.5237	0.4602	0.5455	0.5455	0.5455
0.996	36.0	1044	1.3405	0.4559	0.5530	0.5235	0.4795	0.5530	0.5377	0.4715	0.5530	0.5530	0.5530
0.9613	37.0	1073	1.3357	0.4847	0.5833	0.5535	0.5011	0.5833	0.5612	0.5020	0.5833	0.5833	0.5833
0.8507	38.0	1102	1.3347	0.4760	0.5758	0.5454	0.4897	0.5758	0.5510	0.4940	0.5758	0.5758	0.5758
1.1563	39.0	1131	1.3396	0.4553	0.5530	0.5250	0.4608	0.5530	0.5234	0.4735	0.5530	0.5530	0.5530
0.9681	40.0	1160	1.3371	0.4703	0.5682	0.5396	0.4816	0.5682	0.5445	0.4887	0.5682	0.5682	0.5682

Framework versions

Transformers 4.48.2
Pytorch 2.6.0+cu124
Datasets 3.2.0
Tokenizers 0.21.0

corranm/vit-base-patch16-224-in21k_16batch

vit-base-patch16-224-in21k_16batch