auduno / deepdraw Goto Github PK

Is there some method to support other frameworks?
like Tensorflow, Pytorch?

Hi,
Very cool work! I wonder how feasible it is to generate images with multiple classes together? For instance, "monkey bird dog" or "person car cat"

Imagine the insanity!

Hello! What a nice job! Could you provide code to draw with ResNet? Thank you

Hi,

Have you tried running DeepDraw with other net architectures? For example, I've been able to run it with the ResNet-50 targeting the Softmax layer, but I'm not getting useful visualizations... Any suggestion on how to modify the "octaves" parameters?

Thank you!

Thanks for your great work!
I am new in caffe,recently i am working on the project of "Temporal Segment Networks: Towards Good
Practices for Deep Action Recognition".The author adopt the DeepDraw toolbox to visualize the model.I am stuck in getting the same pic as author's.
I have the files:
1.tsn_bn_inception_flow_deploy.prototxt
2.hmdb51_split_1_tsn_flow_reference_bn_inception.caffemodel
Can you show me how to run the toolbox if i use another model named TSN ?
Thank you very much!