IndoColSmol-256M

This model is a fine-tuned version of vidore/ColSmolVLM-Instruct-256M-base on the ingenio/indodvqa_dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.3786

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0099	1	0.4959
0.4991	0.3960	40	0.4319
0.4293	0.7921	80	0.3986
0.4	1.1881	120	0.3829
0.3653	1.5842	160	0.3788
0.3846	1.9802	200	0.3764

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1