How to deploy? : )

[richdu]

I am running into some issues trying to deploy the code. When I try to deploy the code, the temp_wl and loss1/classifier_wl layers are initialized randomly, so the output is random and doesn't work. As you suggested, I removed the following code:

weight_filler {
  type: "xavier"
}
bias_filler {
  type: "constant"
  value: 0.2
}

But then, all the weights and biases were 0.

Could you advise me on how to properly deploy your pre-trained model? Do I need to modify deploy.prototxt? Currently, I am using deploy.prototxt and ILGnet-AVA1.caffemodel. Your test.py did not seem to work for me.

[BestiVictory]

the file of test.py is updated,you could try again.If this question still exist,please make sure the environment of Caffe is correct.
note:deplory.protxt not need modify

[dangz90]

Hi,

I'm trying to test it on amazon server but I receive the folowing error:
F0920 14:45:31.776003 14806 cudnn_conv_layer.cpp:52] Check failed: error == cudaSuccess (30 vs. 0) unknown error

Do you know if this is an error from the model or the caffe environment?

[BestiVictory]

this error is very common on caffe library, please look help for Google search.

[horiken4]

I've been facing same problem. I'm using test.py and running on caffe official docker image with cpu mode (https://hub.docker.com/r/bvlc/caffe/tags/). But unfortunately, the model outputs a random value in spite of inputting same image.

[dangz90]

@horiken4 How did you run test.py? Everytime I try to run it I receive the following error:
caffe.NetParameter: 6:15: Message type "caffe.LayerParameter" has no field named "input_param".
I'm also using the official docker image but with gpu.