lightonocr-books-test

This model is a fine-tuned version of lightonai/LightOnOCR-1B-1025 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1956

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: 6e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
• lr_scheduler_warmup_steps: 10
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.4379	0.1238	50	0.2103
0.4013	0.2475	100	0.2058
0.3717	0.3713	150	0.1978
0.4181	0.4950	200	0.2009
0.4247	0.6188	250	0.2007
0.3612	0.7426	300	0.1963
0.3554	0.8663	350	0.1971
0.3778	0.9901	400	0.1956

Framework versions

• Transformers 5.0.0.dev0
• Pytorch 2.9.0+cu128
• Datasets 4.4.0
• Tokenizers 0.22.1