mohammadalihumayun/trocr-ur

trocr for Urdu

This model is a fine-tuned version of cxfajar197/urdu-ocr on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0120
• Cer: 0.2500

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.4637	0.3188	1000	2.3244	0.3040
0.4822	0.6376	2000	2.2832	0.3015
0.5518	0.9563	3000	2.0469	0.2796
0.4168	1.2751	4000	2.1507	0.2900
0.375	1.5939	5000	2.0784	0.2744
0.3911	1.9127	6000	2.0120	0.2500

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0