Skip to content

Transfer learning with tensorflow #2

@Ajaz-Ahmad

Description

@Ajaz-Ahmad

Hi,

I am using yolo architecture and trying transfer learning. I have pre-trained weights trained over 80 classes but now I want output for 3 classes only. How can I remove the last convolution layer and fit my own layer?
Below is the initializer method where I am restoring model and training only the last part but I want to change last convolutional layer as well.
Thanks
YOLO_initializer

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions