LiLTInvoiceCzechV0123Test

This model is a fine-tuned version of TomasFAV/LiLTInvoiceCzechV012 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0394
• Precision: 0.8940
• Recall: 0.9215
• F1: 0.9076
• Accuracy: 0.9921

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 2
• 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	12	0.0787	0.7965	0.7679	0.7819	0.9814
No log	2.0	24	0.0585	0.7896	0.8328	0.8106	0.9830
No log	3.0	36	0.0404	0.8902	0.8720	0.8810	0.9894
No log	4.0	48	0.0399	0.8286	0.8908	0.8586	0.9877
No log	5.0	60	0.0357	0.8984	0.8754	0.8868	0.9909
No log	6.0	72	0.0359	0.8709	0.8976	0.8840	0.9905
No log	7.0	84	0.0401	0.8670	0.9232	0.8942	0.9904
No log	8.0	96	0.0388	0.8585	0.9113	0.8841	0.9904
No log	9.0	108	0.0395	0.8891	0.9027	0.8959	0.9907
No log	10.0	120	0.0381	0.885	0.9061	0.8954	0.9914
No log	11.0	132	0.0384	0.8725	0.9113	0.8915	0.9911
No log	12.0	144	0.0400	0.8781	0.9096	0.8935	0.9909
No log	13.0	156	0.0402	0.8918	0.9147	0.9031	0.9919
No log	14.0	168	0.0393	0.8940	0.9215	0.9076	0.9921
No log	15.0	180	0.0415	0.8935	0.9164	0.9048	0.9916
No log	16.0	192	0.0392	0.8961	0.9130	0.9045	0.9919
No log	17.0	204	0.0406	0.8809	0.9215	0.9008	0.9916
No log	18.0	216	0.0414	0.8847	0.9164	0.9003	0.9917
No log	19.0	228	0.0418	0.8865	0.9198	0.9028	0.9918
No log	20.0	240	0.0419	0.8882	0.9215	0.9045	0.9919

Framework versions

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