File size: 11,094 Bytes
714a85e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
import os
import random
import numpy as np
from scipy.ndimage import zoom

# IMPORTANT: paths are 1, walls are 0
# Get all mazes of some size with paths of a given shape with:
# `all_{shape}_path_mazes(size)`

### SQUARES ###

# Return a list of all minimal bounding box squares up to a certain size.
def all_mbbox_squares(max_size):
    all_squares = []
    # Start with smallest possible square.
    curr_square = np.array([[1, 1, 1], [1, 0, 1], [1, 1, 1]]) 
    curr_size = curr_square.shape[0]
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        all_squares.append(curr_square)
        curr_square = zoom(
            curr_square,
            ((curr_size + 1) / curr_size, (curr_size + 1) / curr_size),
            order=0
        )
        curr_size = curr_square.shape[0]
    return all_squares

# Generate all square paths in a maze 
def all_square_path_mazes(maze_size):
    paths = all_mbbox_squares(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### CROSSES ###

def create_cross(size):
    diag_1 = np.eye(size, dtype=int)
    diag_2 = np.fliplr(np.eye(size, dtype=int))
    return diag_1 | diag_2

# Only allowing crosses with an uneven size to avoid a 2x2 block in the middle.
def all_mbbox_crosses(max_size):
    all_crosses = []
    # Start with smallest possible cross.
    curr_size = 3
    curr_cross = create_cross(curr_size)
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        all_crosses.append(curr_cross)
        curr_size += 2
        curr_cross = create_cross(curr_size)
    return all_crosses

def all_cross_path_mazes(maze_size):
    paths = all_mbbox_crosses(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)


### SPIRAL ###

# Fill a square grid with clockwise spiral from top left.
def create_base_spiral(size):
    spiral = np.zeros((size, size), dtype=int)

    top, bottom = 0, size - 1
    left, right = 0, size - 1
    
    while top <= bottom and left <= right:
        # Fill the top row (left to right)
        for i in range(max(left - 1, 0), right + 1):
            spiral[top][i] = 1
        top += 2
        
        # Fill the right column (top to bottom)
        for i in range(top - 1, bottom + 1):
            spiral[i][right] = 1
        right -= 2
        
        if top <= bottom:
            # Fill the bottom row (right to left)
            for i in range(right + 1, left - 1, -1):
                spiral[bottom][i] = 1
            bottom -= 2
        
        if left <= right:
            # Fill the left column (bottom to top)
            for i in range(bottom + 1, top - 1, -1):
                spiral[i][left] = 1
            left += 2
    
    return spiral
 

# Creates all square spirals going both clockwise and anticlockwise in all
# four orientations.
def all_mbbox_spirals(max_size):
    all_spirals = []
    curr_size = 5
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        cw_base_spiral = create_base_spiral(curr_size)
        acw_base_spiral = np.fliplr(cw_base_spiral)
        for k in range(4):
            all_spirals.append(np.rot90(cw_base_spiral, k=k))
            all_spirals.append(np.rot90(acw_base_spiral, k=k))
        curr_size += 1
    return all_spirals

def all_spiral_path_mazes(maze_size):
    paths = all_mbbox_spirals(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### TRIANGLE ###

# Create all triangles with base at bottom, pointing up.
def all_mbbox_bottom_base_triangles(max_size):
    all_triangles = []
    # Smallest base bottom triangle.
    curr_size = (3, 5)
    while curr_size[0] <= max_size[0] and curr_size[1] <= max_size[1]:
        triangle = np.zeros(curr_size, dtype=int)
        # Top row has single 1 in the middle.
        mid_point = curr_size[1] // 2
        triangle[0, mid_point] = 1
        # Bottom row is all 1's.
        triangle[-1, :] = 1
        # Intermediate rows have two 1's expanding out from top to bottom.
        for i in range(1, curr_size[0] - 1):
            triangle[i, mid_point - i] = 1
            triangle[i, mid_point + i] = 1
        all_triangles.append(triangle)
        curr_size = (curr_size[0] + 1, curr_size[1] + 2)
    return all_triangles

# Create all triangles with diagonal bases for all 4 orientations.
def all_mbbox_diag_base_traingles(max_size):
    all_triangles = []
    curr_size = 4
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        triangle = np.eye(curr_size, dtype=int)
        triangle[:, 0] = 1
        triangle[-1, :] = 1
        for k in range(4):
            all_triangles.append(np.rot90(triangle, k=k))
        curr_size += 1
    return all_triangles

def all_triangle_path_mazes(maze_size):
    paths = all_mbbox_bottom_base_triangles(maze_size)
    paths.extend(all_mbbox_diag_base_traingles(maze_size))
    return all_mazes_with_implanted_paths(maze_size, paths)


### C ###

def all_mbbox_Cs(max_size):
    all_cs = []
    # C where length = width.
    curr_c = np.array([[1, 1, 1], [1, 0, 0], [1, 1, 1]]) 
    curr_size = curr_c.shape[0]
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        all_cs.append(curr_c)
        curr_c = zoom(
            curr_c,
            ((curr_size + 1) / curr_size, (curr_size + 1) / curr_size),
            order=0
        )
        curr_size = curr_c.shape[0]
    
    # C where length > width.
    curr_c = np.array([[1, 1, 1], [1, 0, 0], [1, 0, 0], [1, 1, 1]]) 
    while curr_c.shape[0] <= max_size[0] and curr_c.shape[1] <= max_size[1]:
        all_cs.append(curr_c)
        curr_c = zoom(
            curr_c,
            ((curr_c.shape[0] + 1) / curr_c.shape[0], (curr_c.shape[1] + 1) / curr_c.shape[1]),
            order=0
        )

    return all_cs

def all_C_path_mazes(maze_size):
    paths = all_mbbox_Cs(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### U ###

def all_mbbox_Us(max_size):
    all_us = []
    # U where length = width.
    curr_u = np.array([[1, 0, 1], [1, 0, 1], [1, 1, 1]]) 
    curr_size = curr_u.shape[0]
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        all_us.append(curr_u)
        curr_u = zoom(
            curr_u,
            ((curr_size + 1) / curr_size, (curr_size + 1) / curr_size),
            order=0
        )
        curr_size = curr_u.shape[0]
    
    # U where length > width.
    curr_u = np.array([[1, 0, 1], [1, 0, 1], [1, 0, 1], [1, 1, 1]]) 
    while curr_u.shape[0] <= max_size[0] and curr_u.shape[1] <= max_size[1]:
        all_us.append(curr_u)
        curr_u = zoom(
            curr_u,
            ((curr_u.shape[0] + 1) / curr_u.shape[0], (curr_u.shape[1] + 1) / curr_u.shape[1]),
            order=0
        )

    return all_us

def all_U_path_mazes(maze_size):
    paths = all_mbbox_Us(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### Z ###

def create_Z(size):
    z = np.fliplr(np.eye(size, dtype=int))
    z[0, :] = 1
    z[-1, :] = 1
    return z

def all_mbbox_Zs(max_size):
    all_Zs = []
    # Start with smallest possible Z.
    curr_size = 4
    curr_cross = create_Z(curr_size)
    while curr_size <= max_size[0] and curr_size <= max_size[1]:
        all_Zs.append(curr_cross)
        curr_size += 1
        curr_cross = create_Z(curr_size)
    return all_Zs

def all_Z_path_mazes(maze_size):
    paths = all_mbbox_Zs(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### N ###

def all_mbbox_Ns(max_size):
    return [np.rot90(z) for z in all_mbbox_Zs(max_size)]

def all_N_path_mazes(maze_size):
    paths = all_mbbox_Ns(maze_size)
    return all_mazes_with_implanted_paths(maze_size, paths)

### MAZES ###

WALL = 0
PATH = 1
POS = 2
END = 3

MIN_DISTANCE = 2

# Generate all mazes of a given size with each provided path placed into each
# possible position of the maze.
def all_mazes_with_implanted_paths(maze_size, paths):
    all_mazes = []
    for path in paths:
        assert path.shape[0] <= maze_size[0] and path.shape[1] <= maze_size[1]
        for i in range(0, maze_size[0] - path.shape[0] + 1):
            for j in range(0, maze_size[1] - path.shape[1] + 1):
                maze = np.zeros(maze_size, dtype=int)
                maze[i: i + path.shape[0], j: j + path.shape[1]] = path
                all_mazes.append(maze)
    return all_mazes

def distance(p1, p2):
    return np.sqrt((p2[0] - p1[0]) ** 2 + (p2[1] - p1[1]) ** 2)

def init_random_start_end(maze, k=1):
    path_indices = np.argwhere(maze == PATH)
    valid_pairs = []
    for p1 in path_indices:
        for p2 in path_indices:
            if distance(tuple(p1), tuple(p2)) >= MIN_DISTANCE:
                valid_pairs.append((tuple(p1), tuple(p2)))

    random_start_ends = random.sample(valid_pairs, k=k)
    random_mazes = []
    for start, end in random_start_ends:
        random_maze = np.copy(maze)
        random_maze[start] = POS
        random_maze[end] = END
        random_mazes.append(random_maze)

    if k == 1:
        return random_mazes[0]

    return random_mazes

def init_all_start_end(maze):
    path_indices = np.argwhere(maze == PATH)
    valid_pairs = []
    for p1 in path_indices:
        for p2 in path_indices:
            if tuple(p1) != tuple(p2):
                valid_pairs.append((tuple(p1), tuple(p2)))
    mazes = []
    for start, end in valid_pairs:
        path_maze = np.copy(maze)
        path_maze[start] = POS
        path_maze[end] = END
        mazes.append(path_maze)

    return mazes

SHAPES = ["square", "cross", "spiral", "triangle", "C", "Z"]

def get_sample_mazes(maze_size, shape, k):
    if shape not in SHAPES:
        raise NotImplemented("Shape not supported.")
    all_mazes = globals()[f"all_{shape}_path_mazes"](maze_size)
    sample_mazes = random.sample(all_mazes, k=k)
    mazes_with_path = []
    for maze in sample_mazes:
        mazes_with_path.append(init_random_start_end(maze))
    return mazes_with_path

def generate_maze_experiments(size, shape):
    size_shape_path = f"experiment_mazes/{str(size[0])}x{str(size[1])}/{shape}/"
    if not os.path.isdir(size_shape_path):
        os.makedirs(size_shape_path)
    if size == (5, 5):
        # Generate 5 samples of shape
        sample_mazes = get_sample_mazes(size, shape, k=5)
        for i, maze in enumerate(sample_mazes):
            np.save(size_shape_path + f"{str(size[0])}x{str(size[1])}_{shape}_{str(i)}.npy", maze)
    elif size == (7, 7):
        # Generate 30 samples of shape
        sample_mazes = get_sample_mazes(size, shape, k=30)
        for i, maze in enumerate(sample_mazes):
            np.save(size_shape_path + f"{str(size[0])}x{str(size[1])}_{shape}_{str(i)}.npy", maze)
    elif size == (9, 9):
        # Generate 50 samples of shape
        sample_mazes = get_sample_mazes(size, shape, k=50)
        for i, maze in enumerate(sample_mazes):
            np.save(size_shape_path + f"{str(size[0])}x{str(size[1])}_{shape}_{str(i)}.npy", maze)


'''
if __name__ == "__main__":
    for size in [(5, 5), (7, 7), (9, 9)]:
        for shape in SHAPES:
            generate_maze_experiments(size, shape)
'''