TensorRT4:How to do Pooling after "RPROIFused" layer?

Hi ,

I try to do AVE Pooling after RPROIFused and mark the pool6 as output, then I see error message

Begin building engine…
ERROR: cudnnPoolingLayer.cpp (130) - Cudnn Error in execute: 3
ERROR: cudnnPoolingLayer.cpp (130) - Cudnn Error in execute: 3
sample_fasterRCNN: sampleFasterRCNN.cpp:166: void caffeToGIEModel(const string&, const string&, const std::vector<std::basic_string >&, unsigned int, nvcaffeparser1::IPluginFactory*, nvinfer1::IHostMemory**): Assertion `engine’ failed.
Aborted

prototxt:

same as sample - SampleFasterRCNN

layer {
name: “RPROIFused”
type: “IPlugin”
bottom: ‘rpn_cls_prob_reshape’
bottom: ‘rpn_bbox_pred’
bottom: ‘conv5_3’
bottom: ‘im_info’
top: ‘rois’
top: ‘pool5’
}
layer {
bottom: “pool5”
top: “pool6”
name: “pool6”
type: “Pooling”
pooling_param {
kernel_size: 7
stride: 1
pool: AVE
}
}

what should I do for further investigation?

mark

Please file a bug report through our developers portal so our engineers can look into this.