README.md

metadata

library_name: transformers
license: cc-by-nc-sa-4.0
base_model: TomasFAV/Layoutlmv3InvoiceCzechV0
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: Layoutlmv3InvoiceCzechV03
    results: []

Layoutlmv3InvoiceCzechV03

This model is a fine-tuned version of TomasFAV/Layoutlmv3InvoiceCzechV0 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0415
• Precision: 0.8985
• Recall: 0.9289
• F1: 0.9135
• 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: 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	23	0.1122	0.6618	0.7648	0.7096	0.9714
No log	2.0	46	0.0820	0.7325	0.8477	0.7859	0.9793
No log	3.0	69	0.0584	0.8167	0.8596	0.8376	0.9860
No log	4.0	92	0.0536	0.8323	0.9069	0.8680	0.9882
No log	5.0	115	0.0477	0.8390	0.9171	0.8763	0.9890
No log	6.0	138	0.0497	0.8717	0.8849	0.8783	0.9899
No log	7.0	161	0.0424	0.8549	0.9171	0.8849	0.9901
No log	8.0	184	0.0426	0.8728	0.9171	0.8944	0.9911
No log	9.0	207	0.0472	0.8799	0.9052	0.8924	0.9905
No log	10.0	230	0.0471	0.8704	0.9205	0.8947	0.9905
No log	11.0	253	0.0432	0.8860	0.9205	0.9029	0.9913
No log	12.0	276	0.0466	0.8861	0.9086	0.8972	0.9913
No log	13.0	299	0.0438	0.9003	0.9171	0.9086	0.9918
No log	14.0	322	0.0423	0.8831	0.9205	0.9014	0.9914
No log	15.0	345	0.0410	0.8916	0.9188	0.9050	0.9916
No log	16.0	368	0.0448	0.8947	0.9205	0.9074	0.9918
No log	17.0	391	0.0410	0.9010	0.9239	0.9123	0.9921
No log	18.0	414	0.0415	0.8985	0.9289	0.9135	0.9921
No log	19.0	437	0.0425	0.8962	0.9205	0.9082	0.9917
No log	20.0	460	0.0422	0.8962	0.9205	0.9082	0.9918

Framework versions

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