thanks for your code

[c8x1]

it's very helpful to understand the framework of DetNet!
One question is how to use this net as the backbone of yolo, that means the output should be a feature map

[rosivagyok]

@c8x1 Use the output of the last convolutional layer.

[becauseofAI]

@c8x1
I am glad to have been of some help :)
For your question, you can use the pooling or convolution layer to the any layer of detnet to get the feature map you wanted.

[c8x1]

Thank you for your response! It's really helpful:)

…

------------------ Original ------------------ From: becauseofAI <notifications@github.com> Date: Wed,Jun 19,2019 3:00 PM To: becauseofAI/DetNet-Keras <DetNet-Keras@noreply.github.com> Cc: Noc <471565378@qq.com>, Mention <mention@noreply.github.com> Subject: Re: [becauseofAI/DetNet-Keras] thanks for your code (#1) @c8x1 I am glad to have been of some help :) For your question, you can use the pooling or convolution layer to the any layer of detnet to get the feature map you wanted. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.