DertInvoiceCzech / README.md
TomasFAV's picture
Model save
9462eec verified
metadata
library_name: transformers
license: mit
base_model: microsoft/table-transformer-detection
tags:
  - generated_from_trainer
model-index:
  - name: DertInvoiceCzech
    results: []

DertInvoiceCzech

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

  • Loss: 6.0877

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: 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_ratio: 0.1
  • num_epochs: 40
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
No log 1.0 38 8.9100
No log 2.0 76 8.6387
No log 3.0 114 8.2508
No log 4.0 152 7.7641
No log 5.0 190 7.2348
No log 6.0 228 7.0283
No log 7.0 266 6.8791
No log 8.0 304 6.7655
No log 9.0 342 6.6748
No log 10.0 380 6.6063
No log 11.0 418 6.5481
No log 12.0 456 6.4950
No log 13.0 494 6.4460
7.372 14.0 532 6.4030
7.372 15.0 570 6.3710
7.372 16.0 608 6.3404
7.372 17.0 646 6.3093
7.372 18.0 684 6.2916
7.372 19.0 722 6.2786
7.372 20.0 760 6.2660
7.372 21.0 798 6.2440
7.372 22.0 836 6.2129
7.372 23.0 874 6.2026
7.372 24.0 912 6.1897
7.372 25.0 950 6.1778
7.372 26.0 988 6.1668
6.3365 27.0 1026 6.1562
6.3365 28.0 1064 6.1451
6.3365 29.0 1102 6.1313
6.3365 30.0 1140 6.1220
6.3365 31.0 1178 6.1164
6.3365 32.0 1216 6.1105
6.3365 33.0 1254 6.1038
6.3365 34.0 1292 6.1002
6.3365 35.0 1330 6.0966
6.3365 36.0 1368 6.0935
6.3365 37.0 1406 6.0926
6.3365 38.0 1444 6.0905
6.3365 39.0 1482 6.0892
6.1779 40.0 1520 6.0877

Framework versions

  • Transformers 4.57.6
  • Pytorch 2.9.0+cu126
  • Datasets 4.0.0
  • Tokenizers 0.22.2