learning_rate: 0.0001
train_batch_size: 64
eval_batch_size: 64
seed: 42
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: cosine
lr_scheduler_warmup_steps: 150
num_epochs: 100
label_smoothing_factor: 0.15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score	Recall
2.7876	1.0	20	2.7767	0.2340	0.1926	0.2646
2.7688	2.0	40	2.7642	0.2788	0.2392	0.2991
2.7329	3.0	60	2.7404	0.3590	0.3151	0.3345
2.6881	4.0	80	2.7102	0.3878	0.2983	0.3318
2.6213	5.0	100	2.6711	0.4135	0.3370	0.3549
2.5373	6.0	120	2.5738	0.4712	0.4081	0.4215
2.4059	7.0	140	2.4337	0.6058	0.5719	0.5773
2.2029	8.0	160	2.2399	0.6827	0.6608	0.6694
1.9764	9.0	180	1.9648	0.7660	0.7598	0.7581
1.7622	10.0	200	1.7392	0.8109	0.8108	0.8125
1.6197	11.0	220	1.5386	0.8526	0.8520	0.8547
1.4890	12.0	240	1.4154	0.9038	0.9026	0.9063
1.4343	13.0	260	1.3339	0.9135	0.9142	0.9169
1.3343	14.0	280	1.2875	0.9103	0.9113	0.9145
1.3474	15.0	300	1.2767	0.9103	0.9112	0.9157
1.2912	16.0	320	1.2368	0.9263	0.9281	0.9332
1.2806	17.0	340	1.2247	0.9295	0.9309	0.9369
1.2594	18.0	360	1.1994	0.9327	0.9339	0.9394
1.1737	19.0	380	1.1851	0.9359	0.9373	0.9431
1.2589	20.0	400	1.1753	0.9487	0.9504	0.9542
1.1885	21.0	420	1.1794	0.9455	0.9474	0.9518
1.1790	22.0	440	1.1719	0.9519	0.9540	0.9568
1.2309	23.0	460	1.1672	0.9551	0.9575	0.9605
1.1581	24.0	480	1.1660	0.9487	0.9498	0.9524
1.1820	25.0	500	1.1529	0.9551	0.9567	0.9580
1.1390	26.0	520	1.1630	0.9487	0.9498	0.9524
1.1469	27.0	540	1.1615	0.9487	0.9498	0.9524
1.1267	28.0	560	1.1656	0.9487	0.9501	0.9524
1.1421	29.0	580	1.1692	0.9551	0.9566	0.9586
1.1177	30.0	600	1.1556	0.9519	0.9530	0.9549
1.1025	31.0	620	1.1478	0.9551	0.9569	0.9593
1.1242	32.0	640	1.1432	0.9551	0.9570	0.9603
1.0660	33.0	660	1.1552	0.9615	0.9633	0.9655
1.1102	34.0	680	1.1553	0.9583	0.9600	0.9623
1.1102	35.0	700	1.1554	0.9455	0.9465	0.9468
1.0782	36.0	720	1.1303	0.9615	0.9636	0.9661
1.0878	37.0	740	1.1316	0.9647	0.9664	0.9692
1.0976	38.0	760	1.1559	0.9615	0.9632	0.9661
1.0927	39.0	780	1.1404	0.9615	0.9635	0.9667
1.0691	40.0	800	1.1362	0.9583	0.9596	0.9621
1.1037	41.0	820	1.1316	0.9615	0.9629	0.9655
1.0616	42.0	840	1.1195	0.9615	0.9627	0.9642
1.0838	43.0	860	1.1508	0.9615	0.9633	0.9655
1.0782	44.0	880	1.1183	0.9615	0.9624	0.9642
1.0664	45.0	900	1.1364	0.9615	0.9633	0.9655
1.0756	46.0	920	1.1197	0.9583	0.9597	0.9617
1.0538	47.0	940	1.1142	0.9583	0.9592	0.9602
1.0664	48.0	960	1.1452	0.9615	0.9628	0.9652
1.1020	49.0	980	1.1318	0.9583	0.9597	0.9615
1.0452	50.0	1000	1.1231	0.9583	0.9597	0.9615
1.0679	51.0	1020	1.1115	0.9647	0.9662	0.9679
1.0404	52.0	1040	1.1107	0.9615	0.9633	0.9655
1.1221	53.0	1060	1.1152	0.9615	0.9627	0.9642
1.0378	54.0	1080	1.1199	0.9551	0.9562	0.9578
1.0337	55.0	1100	1.1202	0.9551	0.9562	0.9578
1.0463	56.0	1120	1.1344	0.9583	0.9597	0.9615
1.0620	57.0	1140	1.1100	0.9615	0.9633	0.9655
1.0276	58.0	1160	1.1220	0.9583	0.9597	0.9615
1.0399	59.0	1180	1.1143	0.9583	0.9597	0.9615
1.0599	60.0	1200	1.1117	0.9615	0.9627	0.9639
1.0444	61.0	1220	1.1203	0.9583	0.9597	0.9615
1.0342	62.0	1240	1.1150	0.9583	0.9592	0.9602
1.0379	63.0	1260	1.1226	0.9551	0.9562	0.9578
1.0248	64.0	1280	1.1013	0.9647	0.9657	0.9664
1.0181	65.0	1300	1.1058	0.9615	0.9627	0.9639
1.0220	66.0	1320	1.1049	0.9583	0.9592	0.9602
1.0158	67.0	1340	1.1059	0.9615	0.9627	0.9639
1.0248	68.0	1360	1.1040	0.9551	0.9560	0.9565
1.0323	69.0	1380	1.1008	0.9615	0.9627	0.9639
1.0291	70.0	1400	1.0940	0.9647	0.9652	0.9651
1.0199	71.0	1420	1.1207	0.9551	0.9562	0.9578
1.0055	72.0	1440	1.1020	0.9583	0.9592	0.9602
1.0302	73.0	1460	1.1072	0.9583	0.9595	0.9602
1.0322	74.0	1480	1.1047	0.9551	0.9560	0.9565
1.0093	75.0	1500	1.1194	0.9583	0.9597	0.9615
1.0505	76.0	1520	1.0970	0.9615	0.9622	0.9627
1.0396	77.0	1540	1.1056	0.9551	0.9560	0.9565
1.0434	78.0	1560	1.1083	0.9615	0.9627	0.9639
1.0305	79.0	1580	1.1092	0.9615	0.9627	0.9639
1.0298	80.0	1600	1.1054	0.9615	0.9627	0.9639

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

nqvii/resnet50_fold_4_v3

resnet50_fold_4_v3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nqvii/
resnet50_fold_4_v3