Sungchul Kim comments

Results 13 comments of


                                            Sungchul Kim

Enable to load models with low opset version

run tensorflow pre-commit tests

Enable to load models with low opset version

run pytorch pre-commit tests

Enable to load models with low opset version

run pytorch pre-commit tests

Enable to load models with low opset version

@kshpv I saw models from torchvision just have single input, not multiple inputs, and it is not the purpose of this ticket. So, I changed them to models from onnx...

Enable to load models with low opset version

retest this please

Decreased performance with SupCon Loss in Incremental Learning Research Framework

Thanks for your interest, @Daankrol 1. The latter. Unlike the original SupCon, which pre-trains the model with label information, our SupCon targets a fully fine-tuning scheme using both cross-entropy loss...

Decreased performance with SupCon Loss in Incremental Learning Research Framework

@Daankrol Thanks for your ideation! We'll try it and share it if there are notable results :)

OTX v2 visual prompting tasks cannot load datumaro datasets

Could you add `--data.config.data_format=datumaro` in cli?

Instance Segmentation Model refactoring

@kprokofi could you remove the duplicated one? https://github.com/openvinotoolkit/training_extensions/pull/3783#discussion_r1703456339 https://github.com/openvinotoolkit/training_extensions/blob/a5470fc3665228be8c3fde0bdc0728b11b26c057/src/otx/algo/instance_segmentation/layers/transformer.py#L88-L93

Instance Segmentation Model refactoring

@kprokofi Could you check the performance before and after this PR?