CNN之image2column

import numpy as np
import numba


@numba.jit(nopython=False, parallel=True)
def im2col(input_data, filter_h, filter_w, stride=1, pad=0):
    N, C, H, W = input_data.shape
    out_h = (H + 2*pad - filter_h)//stride + 1
    out_w = (W + 2*pad - filter_w)//stride + 1

    img = np.pad(input_data, [(0,0), (0,0), (pad, pad), (pad, pad)], 'constant')
    col = np.zeros((N, C, filter_h, filter_w, out_h, out_w))

    for y in range(filter_h):
        y_max = y + stride*out_h
        for x in range(filter_w):
            x_max = x + stride*out_w
            col[:, :, y, x, :, :] = img[:, :, y:y_max:stride, x:x_max:stride]

    col = col.transpose(0, 4, 5, 1, 2, 3).reshape(N*out_h*out_w, -1)
    return col


img = np.arange(1,65).reshape(2,2,4,4)
print(img)

col = im2col(img, filter_h=1, filter_w=1)
print(col)


'''
[[[[ 1  2  3  4]
   [ 5  6  7  8]
   [ 9 10 11 12]
   [13 14 15 16]]
  [[17 18 19 20]
   [21 22 23 24]
   [25 26 27 28]
   [29 30 31 32]]]
 [[[33 34 35 36]
   [37 38 39 40]
   [41 42 43 44]
   [45 46 47 48]]
  [[49 50 51 52]
   [53 54 55 56]
   [57 58 59 60]
   [61 62 63 64]]]]
[[ 1.  2.  5.  6. 17. 18. 21. 22.]
 [ 2.  3.  6.  7. 18. 19. 22. 23.]
 [ 3.  4.  7.  8. 19. 20. 23. 24.]
 [ 5.  6.  9. 10. 21. 22. 25. 26.]
 [ 6.  7. 10. 11. 22. 23. 26. 27.]
 [ 7.  8. 11. 12. 23. 24. 27. 28.]
 [ 9. 10. 13. 14. 25. 26. 29. 30.]
 [10. 11. 14. 15. 26. 27. 30. 31.]
 [11. 12. 15. 16. 27. 28. 31. 32.]
 [33. 34. 37. 38. 49. 50. 53. 54.]
 [34. 35. 38. 39. 50. 51. 54. 55.]
 [35. 36. 39. 40. 51. 52. 55. 56.]
 [37. 38. 41. 42. 53. 54. 57. 58.]
 [38. 39. 42. 43. 54. 55. 58. 59.]
 [39. 40. 43. 44. 55. 56. 59. 60.]
 [41. 42. 45. 46. 57. 58. 61. 62.]
 [42. 43. 46. 47. 58. 59. 62. 63.]
 [43. 44. 47. 48. 59. 60. 63. 64.]]
'''

class Convolution:
    def __init__(self, W, b, stride=1, pad=0):
        self.W = W
        self.b = b
        self.stride = stride
        self.pad = pad


     def forward(self, x):
         FN, C, FH, FW = self.W.shape
         N, C, H, W = x.shape
         out_h = int(1 + (H + 2*self.pad - FH) / self.stride)
         out_w = int(1 + (W + 2*self.pad - FW) / self.stride)
         col = im2col(x, FH, FW, self.stride, self.pad)
         col_W = self.W.reshape(FN, -1).T # 滤波器的展开
         out = np.dot(col, col_W) + self.b
         out = out.reshape(N, out_h, out_w, -1).transpose(0, 3, 1, 2)
         return out