IshanWarshamana/checkpoints

TrOCR_Printed_Sinahala_1

This model is a fine-tuned version of kavg/TrOCR-SIN-DeiT on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6949
• Cer: 0.1689

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: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.0989	0.0444	100	1.0041	0.1787
0.1265	0.0889	200	0.7479	0.2270
0.0302	0.1333	300	0.9396	0.1863
0.0483	0.1778	400	0.6949	0.1689
0.0637	0.2222	500	0.6022	0.1816
0.0726	0.2667	600	0.6801	0.2083
0.0169	0.3111	700	0.5828	0.1721
0.0084	0.3556	800	0.5437	0.1750
0.0239	0.4	900	0.5451	0.1744
0.0215	0.4444	1000	0.5470	0.1717

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1