Train SSD 600

[paolomanchisi]

HI, I want to train my custom dataset with ssd 600. Is it enough to set

 image_resizer {
      fixed_shape_resizer {
        height: 300
        width: 300
      }
    }

to 600 / 600 instead of 300 / 300 ? Thank you very much.

[gustavz]

yes that is enough, but then you won't be able to use the pre-trained checkpoints. As the weights only fit a model that takes the same input dimensions!

EDIT: Infact you are able to use the pre-trained weights, but you have to set from_detection_checkpoint to false