You can deploy this model and start earning money today!

nehaMe123/
trocr-medical-v2-padded

Name: trocr-medical-v2-padded
Availability: InStock

transformers

safetensors

vision-encoder-decoder

image-text-to-text

generated_from_trainer

microsoft/trocr-small-handwritten

finetune

endpoints_compatible

Views

No views yet

trocr-medical-v2-padded

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

Loss: 0.2892
Cer: 0.3485

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: 4e-05
train_batch_size: 16
eval_batch_size: 16
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
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
2.7747	1.0	166	2.8512	0.8883
2.1227	2.0	332	2.4945	0.8846
1.9645	3.0	498	2.3030	0.8507
1.5736	4.0	664	1.8213	0.8279
1.1897	5.0	830	1.2490	0.7454
0.8294	6.0	996	0.9490	0.5703
0.549	7.0	1162	0.6554	0.5129
0.387	8.0	1328	0.5487	0.4914
0.3109	9.0	1494	0.5268	0.4428
0.2238	10.0	1660	0.4383	0.4076
0.1751	11.0	1826	0.3658	0.3945
0.1141	12.0	1992	0.3795	0.3804
0.0954	13.0	2158	0.3162	0.3640
0.0691	14.0	2324	0.3186	0.3462
0.0589	15.0	2490	0.2892	0.3485

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu126
Datasets 4.4.2
Tokenizers 0.22.1

nehaMe123/trocr-medical-v2-padded

trocr-medical-v2-padded

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nehaMe123/
trocr-medical-v2-padded