Layout-finetuned-fr-model-50instances20-10epochs-5e-05lr-V2

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

• Loss: 0.0001

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: reduce_lr_on_plateau
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
3.2231	0.7692	10	0.7135
0.1961	1.5385	20	0.0019
0.002	2.3077	30	0.0004
0.0005	3.0769	40	0.0002
0.0003	3.8462	50	0.0001
0.0003	4.6154	60	0.0001
0.0002	5.3846	70	0.0001
0.0002	6.1538	80	0.0001
0.0002	6.9231	90	0.0001
0.0002	7.6923	100	0.0001
0.0002	8.4615	110	0.0001
0.0001	9.2308	120	0.0001
0.0001	10.0	130	0.0001

Framework versions

• Transformers 4.48.0
• Pytorch 2.4.1.post100
• Datasets 3.2.0
• Tokenizers 0.21.0