MCPcopy Create free account
hub / github.com/Sin3DM/Sin3DM / forward

Method forward

evaluation/inception.py:113–147  ·  view source on GitHub ↗

Get Inception feature maps Parameters ---------- inp : torch.autograd.Variable Input tensor of shape Bx3xHxW. Values are expected to be in range (0, 1) Returns ------- List of torch.autograd.Variable, corresponding to the sele

(self, inp)

Source from the content-addressed store, hash-verified

111 param.requires_grad = requires_grad
112
113 def forward(self, inp):
114 """Get Inception feature maps
115
116 Parameters
117 ----------
118 inp : torch.autograd.Variable
119 Input tensor of shape Bx3xHxW. Values are expected to be in
120 range (0, 1)
121
122 Returns
123 -------
124 List of torch.autograd.Variable, corresponding to the selected output
125 block, sorted ascending by index
126 """
127 outp = []
128 x = inp
129
130 if self.resize_input:
131 x = F.upsample(x,
132 size=(299, 299),
133 mode='bilinear',
134 align_corners=False)
135
136 if self.normalize_input:
137 x = 2 * x - 1 # Scale from range (0, 1) to range (-1, 1)
138
139 for idx, block in enumerate(self.blocks):
140 x = block(x)
141 if idx in self.output_blocks:
142 outp.append(x)
143
144 if idx == self.last_needed_block:
145 break
146
147 return outp

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected