pytorch-upsampling

comparison.py

import numpy as np
import torch
import torch.nn.functional as F
from torch import nn
from torch.autograd import Variable

def make_bilinear_weights(size, num_channels):
    ''' Make a 2D bilinear kernel suitable for upsampling
    Stack the bilinear kernel for application to tensor '''
    factor = (size + 1) // 2
    if size % 2 == 1:
        center = factor - 1
    else:
        center = factor - 0.5
    og = np.ogrid[:size, :size]
    filt = (1 - abs(og[0] - center) / factor) * \
             (1 - abs(og[1] - center) / factor)
    print filt
    filt = torch.from_numpy(filt)
    w = torch.zeros(num_channels, num_channels, size, size)
    for i in range(num_channels):
        w[i, i] = filt
    return w
    
# Define a toy grid
x = np.array([[1,2],[3,4]], dtype=np.float32)
x = Variable(torch.from_numpy(x[np.newaxis, np.newaxis, :,:]))

# Upsample using Pytorch bilinear upsampling
out1 = F.upsample(x, None, 2, 'bilinear')

# Upsample using transposed convolution
# kernel size is 2x the upsample rate for smoothing
# output will need to be cropped to size
out2 = F.conv_transpose2d(x, Variable(make_bilinear_weights(4, 1)), stride=2)

''' 
Output 
out1 = [ 1.          1.33333325  1.66666675  2.        ]
   [ 1.66666663  2.          2.33333349  2.66666651]
   [ 2.33333325  2.66666675  3.00000024  3.33333349]
   [ 3.          3.33333325  3.66666675  4.        ]
   
out2 = [ 0.0625  0.1875  0.3125  0.4375  0.375   0.125 ]
   [ 0.1875  0.5625  0.9375  1.3125  1.125   0.375 ]
   [ 0.375   1.125   1.75    2.25    1.875   0.625 ]
   [ 0.625   1.875   2.75    3.25    2.625   0.875 ]
   [ 0.5625  1.6875  2.4375  2.8125  2.25    0.75  ]
   [ 0.1875  0.5625  0.8125  0.9375  0.75    0.25  ]
'''