Layoutlmv3InvoiceCzechV3

This model is a fine-tuned version of microsoft/layoutlmv3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0874
• Precision: 0.6694
• Recall: 0.6920
• F1: 0.6805
• Accuracy: 0.9804

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: 1e-05
• train_batch_size: 8
• 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
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	23	2.5446	0.0	0.0	0.0	0.9472
No log	2.0	46	0.8320	0.0	0.0	0.0	0.9473
No log	3.0	69	0.4185	0.0	0.0	0.0	0.9473
No log	4.0	92	0.3832	0.0	0.0	0.0	0.9473
No log	5.0	115	0.2963	0.0	0.0	0.0	0.9473
No log	6.0	138	0.2591	0.0357	0.0017	0.0032	0.9473
No log	7.0	161	0.2357	0.2468	0.1320	0.1720	0.9510
No log	8.0	184	0.2226	0.4192	0.2589	0.3201	0.9574
No log	9.0	207	0.2062	0.5011	0.3875	0.4370	0.9633
No log	10.0	230	0.1946	0.5164	0.4264	0.4671	0.9651
No log	11.0	253	0.1839	0.5515	0.4349	0.4863	0.9663
No log	12.0	276	0.1724	0.5376	0.4839	0.5093	0.9677
No log	13.0	299	0.1675	0.5824	0.5381	0.5594	0.9699
No log	14.0	322	0.1569	0.6127	0.5567	0.5833	0.9709
No log	15.0	345	0.1298	0.6084	0.5888	0.5985	0.9719
No log	16.0	368	0.1226	0.5652	0.5939	0.5792	0.9729
No log	17.0	391	0.1157	0.5621	0.5973	0.5792	0.9739
No log	18.0	414	0.1148	0.5863	0.6210	0.6031	0.9757
No log	19.0	437	0.1134	0.5974	0.6176	0.6073	0.9760
No log	20.0	460	0.1093	0.5866	0.6244	0.6049	0.9757
No log	21.0	483	0.1030	0.5953	0.6396	0.6166	0.9772
0.4082	22.0	506	0.1027	0.6025	0.6413	0.6213	0.9771
0.4082	23.0	529	0.1017	0.6093	0.6464	0.6273	0.9776
0.4082	24.0	552	0.1049	0.6104	0.6362	0.6230	0.9773
0.4082	25.0	575	0.0970	0.5913	0.6413	0.6153	0.9767
0.4082	26.0	598	0.0922	0.6069	0.6582	0.6315	0.9777
0.4082	27.0	621	0.0937	0.6154	0.6633	0.6384	0.9782
0.4082	28.0	644	0.0934	0.6266	0.6616	0.6436	0.9787
0.4082	29.0	667	0.0921	0.6177	0.6616	0.6389	0.9785
0.4082	30.0	690	0.0904	0.6109	0.6616	0.6353	0.9783
0.4082	31.0	713	0.0922	0.6194	0.6582	0.6382	0.9786
0.4082	32.0	736	0.0896	0.6304	0.6667	0.6480	0.9791
0.4082	33.0	759	0.0903	0.6314	0.6667	0.6486	0.9793
0.4082	34.0	782	0.0879	0.6377	0.6819	0.6590	0.9794
0.4082	35.0	805	0.0863	0.6439	0.6853	0.6639	0.9798
0.4082	36.0	828	0.0860	0.6421	0.6768	0.6590	0.9794
0.4082	37.0	851	0.0874	0.6721	0.6937	0.6828	0.9805
0.4082	38.0	874	0.0861	0.6559	0.6870	0.6711	0.9799
0.4082	39.0	897	0.0867	0.6694	0.6920	0.6805	0.9803
0.4082	40.0	920	0.0860	0.6667	0.6937	0.6799	0.9803

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2