IndoColSmol-500M

This model is a fine-tuned version of vidore/ColSmolVLM-Instruct-500M-base on the ingenio/indodvqa_dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.3641

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	0.4474
0.4523	0.3960	40	0.4055
0.3996	0.7921	80	0.3804
0.3637	1.1881	120	0.3687
0.345	1.5842	160	0.3627
0.3466	1.9802	200	0.3630

Framework versions

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