detr_finetuned

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

• Loss: 4914.5493

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: 2
• eval_batch_size: 2
• 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: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
4887.0844	0.2381	100	4915.8008
4806.7622	0.4762	200	4915.0762
4883.3641	0.7143	300	4914.7651
4869.7525	0.9524	400	4914.6387
4809.9978	1.1905	500	4914.6133
4820.81	1.4286	600	4914.6948
4779.5872	1.6667	700	4914.6743
5094.1991	1.9048	800	4914.5752
4851.7441	2.1429	900	4914.6494
4928.8484	2.3810	1000	4914.5767
4852.6178	2.6190	1100	4914.5840
4855.8131	2.8571	1200	4914.5991
4948.5747	3.0952	1300	4914.5967
4887.945	3.3333	1400	4914.5645
4900.1669	3.5714	1500	4914.5747
4937.1328	3.8095	1600	4914.5571
4792.3219	4.0476	1700	4914.6787
4842.8072	4.2857	1800	4914.5640
4914.0503	4.5238	1900	4914.6113
4892.0153	4.7619	2000	4914.5693
4882.0288	5.0	2100	4914.5630
4903.9891	5.2381	2200	4914.5679
4870.5566	5.4762	2300	4914.5688
4919.3287	5.7143	2400	4914.5508
4927.9272	5.9524	2500	4914.5488
4981.8925	6.1905	2600	4914.5537
4864.6322	6.4286	2700	4914.5835
4794.4006	6.6667	2800	4914.5820
4878.885	6.9048	2900	4914.5488
4967.0887	7.1429	3000	4914.5518
4937.0766	7.3810	3100	4914.5464
4829.3891	7.6190	3200	4914.5493
4812.0778	7.8571	3300	4914.5459
4823.5034	8.0952	3400	4914.5444
4919.2544	8.3333	3500	4914.5474
4838.375	8.5714	3600	4914.5581
4832.6153	8.8095	3700	4914.5513
4787.5813	9.0476	3800	4914.5464
4862.2234	9.2857	3900	4914.5464
4878.2669	9.5238	4000	4914.5474
4933.3856	9.7619	4100	4914.5488
4945.8159	10.0	4200	4914.5493

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Tokenizers 0.21.0