feevos removed a comment on issue #9288: Get HybridBlock layer shape on runtime URL: https://github.com/apache/incubator-mxnet/issues/9288#issuecomment-409782762 PS And this is an (hybrid-able) implementation of the pyramid scene parsing module, users need to write their own Conv2DNormed HybridBlock layer (basically, convolution + NormLayer (Batch/Instance etc). Here I assume that input is the format (Batch,NChannels, H, W), with H == W. Trivial to modify for H != W ```Python from mxnet import gluon from mxnet.gluon import HybridBlock from mxnet.ndarray import NDArray from phaino.nn.layers.conv2Dnormed import * # You need to define your own conv2Dnormed HybridBlock class PSP_Pooling(HybridBlock): """ Pyramid Scene Parsing pooling layer, as defined in Zhao et al. 2017 (https://arxiv.org/abs/1612.01105) This is only the pyramid pooling module. INPUT: layer of size Nbatch, Nchannel, H, W OUTPUT: layer of size Nbatch, Nchannel, H, W. """ def __init__(self, _nfilters, _norm_type = 'BatchNorm', **kwards): HybridBlock.__init__(self,**kwards) self.nfilters = _nfilters self.layer_size = None # This is used as a container (list) of layers self.convs = gluon.nn.HybridSequential() with self.name_scope(): self.convs.add(Conv2DNormed(self.nfilters//4,kernel_size=(1,1),padding=(0,0), prefix="_conv1_")) self.convs.add(Conv2DNormed(self.nfilters//4,kernel_size=(1,1),padding=(0,0), prefix="_conv2_")) self.convs.add(Conv2DNormed(self.nfilters//4,kernel_size=(1,1),padding=(0,0), prefix="_conv3_")) self.convs.add(Conv2DNormed(self.nfilters//4,kernel_size=(1,1),padding=(0,0), prefix="_conv4_")) self.conv_norm_final = Conv2DNormed(channels = self.nfilters, kernel_size=(1,1), padding=(0,0), _norm_type=_norm_type) def forward(self,_input): self.layer_size = _input.shape return HybridBlock.forward(self,_input) def hybrid_forward(self,F,_input): p = [_input] for i in range(4): pool_size = self.layer_size[-1] // (2**i) # Need this to be integer x = F.Pooling(_input,kernel=[pool_size,pool_size],stride=[pool_size,pool_size],pool_type='max') x = F.UpSampling(x,sample_type='nearest',scale=pool_size) x = self.convs[i](x) p += [x] out = F.concat(p[0],p[1],p[2],p[3],p[4],dim=1) out = self.conv_norm_final(out) return out ```
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
