learning_rate: 1e-05
train_batch_size: 1
eval_batch_size: 1
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: linear
num_epochs: 101
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Token Accuracy
7.7171	1.0	28	7.5592	0.0069	0.1071
4.76	2.0	56	4.5676	0.0069	0.1429
3.9683	3.0	84	3.7493	0.0069	0.1905
3.6926	4.0	112	3.3863	0.0069	0.1786
3.3937	5.0	140	3.1569	0.0069	0.2143
3.2322	6.0	168	3.0481	0.0069	0.2381
3.1991	7.0	196	2.9807	0.0069	0.2024
3.227	8.0	224	2.9449	0.0069	0.2381
3.1912	9.0	252	2.8770	0.0069	0.2262
2.9805	10.0	280	2.8261	0.0069	0.2381
3.0095	11.0	308	2.7662	0.0069	0.2262
2.9173	12.0	336	2.7037	0.0069	0.2262
2.8839	13.0	364	2.6239	0.0069	0.1667
2.7696	14.0	392	2.5674	0.0069	0.2143
2.7105	15.0	420	2.4985	0.0069	0.1667
2.6607	16.0	448	2.4279	0.0069	0.1905
2.3572	17.0	476	2.4473	0.0069	0.1905
2.6981	18.0	504	2.3068	0.0069	0.1905
2.5758	19.0	532	2.2849	0.0069	0.1786
2.525	20.0	560	2.2139	0.0069	0.2262
2.4038	21.0	588	2.1885	0.0069	0.1905
2.0574	22.0	616	2.1600	0.0069	0.1667
2.3321	23.0	644	2.1691	0.0069	0.1667
2.226	24.0	672	2.0827	0.0069	0.2024
2.3477	25.0	700	2.0351	0.0069	0.1667
2.3009	26.0	728	2.0007	0.0069	0.1905
1.9822	27.0	756	1.9625	0.0069	0.1548
2.2624	28.0	784	1.9249	0.0069	0.2619
2.195	29.0	812	1.8983	0.0069	0.1905
2.1884	30.0	840	1.8731	0.0069	0.2024
2.2705	31.0	868	1.8565	0.0069	0.1905
1.6815	32.0	896	1.8052	0.0069	0.2262
1.6833	33.0	924	1.8118	0.0069	0.1667
2.0172	34.0	952	1.7668	0.0069	0.2024
1.7682	35.0	980	1.7230	0.0069	0.1905
2.0052	36.0	1008	1.7116	0.0069	0.2381
1.9945	37.0	1036	1.6614	0.0069	0.1667
1.8177	38.0	1064	1.6244	0.0069	0.1905
1.6026	39.0	1092	1.6156	0.0069	0.2024
1.698	40.0	1120	1.5931	0.0069	0.1548
1.958	41.0	1148	1.5904	0.0069	0.1905
1.81	42.0	1176	1.5726	0.0069	0.1905
1.8077	43.0	1204	1.5334	0.0069	0.1786
1.5514	44.0	1232	1.5127	0.0069	0.1905
1.6704	45.0	1260	1.4878	0.0069	0.1905
1.3608	46.0	1288	1.4477	0.0069	0.2024
1.8209	47.0	1316	1.4363	0.0069	0.1905
1.5357	48.0	1344	1.4245	0.0069	0.1786
1.2907	49.0	1372	1.4105	0.0069	0.1905
1.1673	50.0	1400	1.4127	0.0069	0.2143
1.4882	51.0	1428	1.3726	0.0069	0.1786
1.443	52.0	1456	1.3527	0.0069	0.1667
1.5143	53.0	1484	1.3622	0.0069	0.2024
1.5727	54.0	1512	1.3343	0.0069	0.2024
1.5824	55.0	1540	1.2902	0.0069	0.1905
1.2832	56.0	1568	1.2859	0.0069	0.1905
1.3842	57.0	1596	1.2868	0.0069	0.1667
1.0809	58.0	1624	1.2630	0.0069	0.1786
1.0325	59.0	1652	1.2616	0.0069	0.1786
1.0043	60.0	1680	1.2382	0.0069	0.2143
1.1068	61.0	1708	1.2175	0.0069	0.1905
1.4251	62.0	1736	1.1905	0.0069	0.1667
1.0549	63.0	1764	1.1730	0.0069	0.1548
1.0225	64.0	1792	1.1767	0.0069	0.1786
1.2721	65.0	1820	1.1618	0.0069	0.1667
0.881	66.0	1848	1.1818	0.0069	0.1429
1.0059	67.0	1876	1.1653	0.0069	0.1786
1.3893	68.0	1904	1.1454	0.0069	0.1548
0.8277	69.0	1932	1.1410	0.0069	0.1548
0.9612	70.0	1960	1.1184	0.0069	0.1429
1.3136	71.0	1988	1.0942	0.0069	0.1548
1.2892	72.0	2016	1.0975	0.0069	0.1905
1.3924	73.0	2044	1.0884	0.0069	0.1667
0.8638	74.0	2072	1.0778	0.0069	0.1667
1.0998	75.0	2100	1.0737	0.0069	0.1786
1.0305	76.0	2128	1.0775	0.0069	0.1786
0.9442	77.0	2156	1.0699	0.0069	0.1786
1.4239	78.0	2184	1.0699	0.0069	0.1667
1.1878	79.0	2212	1.0614	0.0069	0.2024
1.085	80.0	2240	1.0577	0.0069	0.1905
0.7415	81.0	2268	1.0510	0.0069	0.1786
0.6056	82.0	2296	1.0418	0.0069	0.1905
0.9905	83.0	2324	1.0365	0.0069	0.1548
1.1389	84.0	2352	1.0417	0.0069	0.1905
0.5456	85.0	2380	1.0341	0.0069	0.1905
1.1479	86.0	2408	1.0267	0.0069	0.1786
0.8913	87.0	2436	1.0108	0.0069	0.1905
1.3453	88.0	2464	1.0162	0.0069	0.1786
1.0019	89.0	2492	1.0156	0.0069	0.1786
0.9211	90.0	2520	1.0206	0.0069	0.1786
1.0389	91.0	2548	1.0187	0.0069	0.1905
1.3368	92.0	2576	1.0134	0.0069	0.1905
1.0589	93.0	2604	1.0143	0.0069	0.1905
0.9895	94.0	2632	1.0107	0.0069	0.1905
0.9452	95.0	2660	1.0093	0.0069	0.1905
0.8314	96.0	2688	1.0060	0.0069	0.1905
0.8572	97.0	2716	1.0045	0.0069	0.1905
0.7942	98.0	2744	1.0030	0.0069	0.1905
0.9418	99.0	2772	1.0024	0.0069	0.1905
1.3698	100.0	2800	1.0029	0.0069	0.1905
1.5673	101.0	2828	0.3344	0.0049	0.1786

Framework versions

Transformers 4.56.2
Pytorch 2.8.0+cu128
Datasets 4.1.1
Tokenizers 0.22.1

Cyber-Zhaba/donut-metric-cells

donut-metric-cells

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Cyber-Zhaba/
donut-metric-cells