LightOnOCR-2-ft-01 – AI Model by Phu-Hien | AlphaNeural AI

You can deploy this model and start earning money today!

Phu-Hien/
LightOnOCR-2-ft-01

like

0

text-generation

lightonai/LightOnOCR-2-1B-base

Views

No views yet

LightOnOCR-2-ft-01

This model is a fine-tuned version of lightonai/LightOnOCR-2-1B-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0613

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: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 4
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
0.1561	0.1074	50	0.1297
0.1339	0.2148	100	0.0887
0.0739	0.3222	150	0.0774
0.0615	0.4296	200	0.0714
0.0698	0.5371	250	0.0678
0.0716	0.6445	300	0.0646
0.0589	0.7519	350	0.0628
0.0563	0.8593	400	0.0616
0.0737	0.9667	450	0.0613

Framework versions

PEFT 0.19.1
Transformers 5.0.0
Pytorch 2.11.0+cu128
Datasets 4.8.5
Tokenizers 0.22.2