【400】numpy.pad 为数组加垫(迷宫类题目)

参考：Numpy学习——数组填充np.pad()函数的应用

---

 

举例说明：

?

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 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

import numpy as np a = np.zeros((3, 4), dtype=int) a   array([[0, 0, 0, 0],        [0, 0, 0, 0],        [0, 0, 0, 0]])   # pad(array, pad_width, mode, **kwargs) # array: array_like # pad_width: ((1, 2), (3, 4)) #    1: width of top #    2: width of bottom #    3: width of left #    4: width of right # mode: str or function #    'constant': Pads with a constant value # constant_values: Used in 'constant'. The values to set the padded values for each axis. #    ((1, 2), (3, 4)) #    1: value of top #    2: value of bottom #    3: vlaue of left #    4: vlaue of right np.pad(a, ((1, 2), (3, 4)), 'constant', constant_values=((1, 2), (3, 4)))   array([[3, 3, 3, 1, 1, 1, 1, 4, 4, 4, 4],        [3, 3, 3, 0, 0, 0, 0, 4, 4, 4, 4],        [3, 3, 3, 0, 0, 0, 0, 4, 4, 4, 4],        [3, 3, 3, 0, 0, 0, 0, 4, 4, 4, 4],        [3, 3, 3, 2, 2, 2, 2, 4, 4, 4, 4],        [3, 3, 3, 2, 2, 2, 2, 4, 4, 4, 4]])   # (1, 2) # top:bottom=1:2 # left:right=1:2 np.pad(a, (1, 2), 'constant', constant_values=(1, 2))   array([[1, 1, 1, 1, 1, 2, 2],        [1, 0, 0, 0, 0, 2, 2],        [1, 0, 0, 0, 0, 2, 2],        [1, 0, 0, 0, 0, 2, 2],        [1, 2, 2, 2, 2, 2, 2],        [1, 2, 2, 2, 2, 2, 2]])   a = [1, 2, 3, 4, 5] np.pad(a, (2, 3), 'constant', constant_values=(4, 6))   array([4, 4, 1, 2, 3, 4, 5, 6, 6, 6])   # mode: 'edge': Pads with the edge values of array np.pad(a, (2, 3), 'edge')   array([1, 1, 1, 2, 3, 4, 5, 5, 5, 5])   np.pad(a, (2, 3), 'linear_ramp', end_values=(5, 1))   array([5, 3, 1, 2, 3, 4, 5, 4, 2, 1])   np.pad(a, (2, 3), 'maximum')   array([5, 5, 1, 2, 3, 4, 5, 5, 5, 5])   np.pad(a, (2, 3), 'mean')   array([3, 3, 1, 2, 3, 4, 5, 3, 3, 3])   np.pad(a, (2, 3), 'median')   array([3, 3, 1, 2, 3, 4, 5, 3, 3, 3])   a = [[1, 2], [3, 4]] np.pad(a, (2, 3), 'minimum')   array([[1, 1, 1, 2, 1, 1, 1],        [1, 1, 1, 2, 1, 1, 1],        [1, 1, 1, 2, 1, 1, 1],        [3, 3, 3, 4, 3, 3, 3],        [1, 1, 1, 2, 1, 1, 1],        [1, 1, 1, 2, 1, 1, 1],        [1, 1, 1, 2, 1, 1, 1]])

---

语法说明：

?

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 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204

>>> help(np.pad) Help on function pad in module numpy.lib.arraypad:   pad(array, pad_width, mode, **kwargs)     Pads an array.           Parameters     ----------     array : array_like of rank N         Input array     pad_width : {sequence, array_like, int}         Number of values padded to the edges of each axis.         ((before_1, after_1), ... (before_N, after_N)) unique pad widths         for each axis.         ((before, after),) yields same before and after pad for each axis.         (pad,) or int is a shortcut for before = after = pad width for all         axes.     mode : str or function         One of the following string values or a user supplied function.               'constant'             Pads with a constant value.         'edge'             Pads with the edge values of array.         'linear_ramp'             Pads with the linear ramp between end_value and the             array edge value.         'maximum'             Pads with the maximum value of all or part of the             vector along each axis.         'mean'             Pads with the mean value of all or part of the             vector along each axis.         'median'             Pads with the median value of all or part of the             vector along each axis.         'minimum'             Pads with the minimum value of all or part of the             vector along each axis.         'reflect'             Pads with the reflection of the vector mirrored on             the first and last values of the vector along each             axis.         'symmetric'             Pads with the reflection of the vector mirrored             along the edge of the array.         'wrap'             Pads with the wrap of the vector along the axis.             The first values are used to pad the end and the             end values are used to pad the beginning.         <function>             Padding function, see Notes.     stat_length : sequence or int, optional         Used in 'maximum', 'mean', 'median', and 'minimum'.  Number of         values at edge of each axis used to calculate the statistic value.               ((before_1, after_1), ... (before_N, after_N)) unique statistic         lengths for each axis.               ((before, after),) yields same before and after statistic lengths         for each axis.               (stat_length,) or int is a shortcut for before = after = statistic         length for all axes.               Default is ``None``, to use the entire axis.     constant_values : sequence or int, optional         Used in 'constant'.  The values to set the padded values for each         axis.               ((before_1, after_1), ... (before_N, after_N)) unique pad constants         for each axis.               ((before, after),) yields same before and after constants for each         axis.               (constant,) or int is a shortcut for before = after = constant for         all axes.               Default is 0.     end_values : sequence or int, optional         Used in 'linear_ramp'.  The values used for the ending value of the         linear_ramp and that will form the edge of the padded array.               ((before_1, after_1), ... (before_N, after_N)) unique end values         for each axis.               ((before, after),) yields same before and after end values for each         axis.               (constant,) or int is a shortcut for before = after = end value for         all axes.               Default is 0.     reflect_type : {'even', 'odd'}, optional         Used in 'reflect', and 'symmetric'.  The 'even' style is the         default with an unaltered reflection around the edge value.  For         the 'odd' style, the extended part of the array is created by         subtracting the reflected values from two times the edge value.           Returns     -------     pad : ndarray         Padded array of rank equal to `array` with shape increased         according to `pad_width`.           Notes     -----     .. versionadded:: 1.7.0           For an array with rank greater than 1, some of the padding of later     axes is calculated from padding of previous axes.  This is easiest to     think about with a rank 2 array where the corners of the padded array     are calculated by using padded values from the first axis.           The padding function, if used, should return a rank 1 array equal in     length to the vector argument with padded values replaced. It has the     following signature::               padding_func(vector, iaxis_pad_width, iaxis, kwargs)           where               vector : ndarray             A rank 1 array already padded with zeros.  Padded values are             vector[:pad_tuple[0]] and vector[-pad_tuple[1]:].         iaxis_pad_width : tuple             A 2-tuple of ints, iaxis_pad_width[0] represents the number of             values padded at the beginning of vector where             iaxis_pad_width[1] represents the number of values padded at             the end of vector.         iaxis : int             The axis currently being calculated.         kwargs : dict             Any keyword arguments the function requires.           Examples     --------     >>> a = [1, 2, 3, 4, 5]     >>> np.pad(a, (2,3), 'constant', constant_values=(4, 6))     array([4, 4, 1, 2, 3, 4, 5, 6, 6, 6])           >>> np.pad(a, (2, 3), 'edge')     array([1, 1, 1, 2, 3, 4, 5, 5, 5, 5])           >>> np.pad(a, (2, 3), 'linear_ramp', end_values=(5, -4))     array([ 5,  3,  1,  2,  3,  4,  5,  2, -1, -4])           >>> np.pad(a, (2,), 'maximum')     array([5, 5, 1, 2, 3, 4, 5, 5, 5])           >>> np.pad(a, (2,), 'mean')     array([3, 3, 1, 2, 3, 4, 5, 3, 3])           >>> np.pad(a, (2,), 'median')     array([3, 3, 1, 2, 3, 4, 5, 3, 3])           >>> a = [[1, 2], [3, 4]]     >>> np.pad(a, ((3, 2), (2, 3)), 'minimum')     array([[1, 1, 1, 2, 1, 1, 1],            [1, 1, 1, 2, 1, 1, 1],            [1, 1, 1, 2, 1, 1, 1],            [1, 1, 1, 2, 1, 1, 1],            [3, 3, 3, 4, 3, 3, 3],            [1, 1, 1, 2, 1, 1, 1],            [1, 1, 1, 2, 1, 1, 1]])           >>> a = [1, 2, 3, 4, 5]     >>> np.pad(a, (2, 3), 'reflect')     array([3, 2, 1, 2, 3, 4, 5, 4, 3, 2])           >>> np.pad(a, (2, 3), 'reflect', reflect_type='odd')     array([-1,  0,  1,  2,  3,  4,  5,  6,  7,  8])           >>> np.pad(a, (2, 3), 'symmetric')     array([2, 1, 1, 2, 3, 4, 5, 5, 4, 3])           >>> np.pad(a, (2, 3), 'symmetric', reflect_type='odd')     array([0, 1, 1, 2, 3, 4, 5, 5, 6, 7])           >>> np.pad(a, (2, 3), 'wrap')     array([4, 5, 1, 2, 3, 4, 5, 1, 2, 3])           >>> def pad_with(vector, pad_width, iaxis, kwargs):     ...     pad_value = kwargs.get('padder', 10)     ...     vector[:pad_width[0]] = pad_value     ...     vector[-pad_width[1]:] = pad_value     ...     return vector     >>> a = np.arange(6)     >>> a = a.reshape((2, 3))     >>> np.pad(a, 2, pad_with)     array([[10, 10, 10, 10, 10, 10, 10],            [10, 10, 10, 10, 10, 10, 10],            [10, 10,  0,  1,  2, 10, 10],            [10, 10,  3,  4,  5, 10, 10],            [10, 10, 10, 10, 10, 10, 10],            [10, 10, 10, 10, 10, 10, 10]])     >>> np.pad(a, 2, pad_with, padder=100)     array([[100, 100, 100, 100, 100, 100, 100],            [100, 100, 100, 100, 100, 100, 100],            [100, 100,   0,   1,   2, 100, 100],            [100, 100,   3,   4,   5, 100, 100],            [100, 100, 100, 100, 100, 100, 100],            [100, 100, 100, 100, 100, 100, 100]])