def forward()

in utils/model.py [0:0]

• 10 lines of code
• 1 McCabe index (conditional complexity)

	def forward(self, features, poses, homography, poses2, lengths):
		""" decode image feature vectors and generate pose sequences """
		poses = self.embed(poses)
		poses[:, 0, :] = torch.zeros([poses.shape[0], self.embed_size]).cuda().float()

		# concat the embedding with (im features, homographies)
		embeddings = torch.cat((poses, features), 2)
		embeddings = torch.cat((features, homography.cuda()), 2)
		embeddings = torch.cat((embeddings, poses2.cuda()), 2)
		packed = pack_padded_sequence(embeddings, lengths, batch_first=True)
		hiddens, _ = self.lstm(packed)

		# transform result to size of vocab (each word has score)
		outputs = self.linear(hiddens[0])
		return outputs