about summary refs log tree commit diff
path: root/users/wpcarro/scratch/deepmind/part_two/misc
diff options
context:
space:
mode:
Diffstat (limited to 'users/wpcarro/scratch/deepmind/part_two/misc')
-rw-r--r--users/wpcarro/scratch/deepmind/part_two/misc/matrix-traversals.py104
1 files changed, 104 insertions, 0 deletions
diff --git a/users/wpcarro/scratch/deepmind/part_two/misc/matrix-traversals.py b/users/wpcarro/scratch/deepmind/part_two/misc/matrix-traversals.py
new file mode 100644
index 000000000000..52354f990e11
--- /dev/null
+++ b/users/wpcarro/scratch/deepmind/part_two/misc/matrix-traversals.py
@@ -0,0 +1,104 @@
+# Herein I'm practicing two-dimensional matrix traversals in all directions of
+# which I can conceive:
+# 0. T -> B; L -> R
+# 1. T -> B; R -> L
+# 2. B -> T; L -> R
+# 3. B -> T; R -> L
+#
+# Commentary:
+# When I think of matrices, I'm reminded of cartesian planes. I think of the
+# cells as (X,Y) coordinates. This has been a pitfall for me because matrices
+# are usually encoded in the opposite way. That is, to access a cell at the
+# coordinates (X,Y) given a matrix M, you index M like this: M[Y][X]. To attempt
+# to avoid this confusion, instead of saying X and Y, I will prefer saying
+# "column" and "row".
+#
+# When traversing a matrix, you typically traverse vertically and then
+# horizontally; in other words, the rows come first followed by the columns. As
+# such, I'd like to refer to traversal orders as "top-to-bottom, left-to-right"
+# rather than "left-to-right, top-to-bottom".
+#
+# These practices are all in an attempt to rewire my thinking.
+
+# This is a list of matrices where the index of a matrix corresponds to the
+# order in which it should be traversed to produce the sequence:
+# [1,2,3,4,5,6,7,8,9].
+boards = [[[1, 2, 3], [4, 5, 6], [7, 8, 9]], [[3, 2, 1], [6, 5, 4], [9, 8, 7]],
+          [[7, 8, 9], [4, 5, 6], [1, 2, 3]], [[9, 8, 7], [6, 5, 4], [3, 2, 1]]]
+
+# T -> B; L -> R
+board = boards[0]
+result = []
+for row in board:
+    for col in row:
+        result.append(col)
+print(result)
+
+# T -> B; R -> L
+board = boards[1]
+result = []
+for row in board:
+    for col in reversed(row):
+        result.append(col)
+print(result)
+
+# B -> T; L -> R
+board = boards[2]
+result = []
+for row in reversed(board):
+    for col in row:
+        result.append(col)
+print(result)
+
+# B -> T; R -> L
+board = boards[3]
+result = []
+for row in reversed(board):
+    for col in reversed(row):
+        result.append(col)
+print(result)
+
+################################################################################
+# Neighbors
+################################################################################
+
+import random
+
+
+# Generate a matrix of size `rows` x `cols` where each cell contains an item
+# randomly selected from `xs`.
+def generate_board(rows, cols, xs):
+    result = []
+    for _ in range(rows):
+        row = []
+        for _ in range(cols):
+            row.append(random.choice(xs))
+        result.append(row)
+    return result
+
+
+# Print the `board` to the screen.
+def print_board(board):
+    print('\n'.join([' '.join(row) for row in board]))
+
+
+board = generate_board(4, 5, ['R', 'G', 'B'])
+print_board(board)
+
+
+# Return all of the cells horizontally and vertically accessible from a starting
+# cell at `row`, `col` in `board`.
+def neighbors(row, col, board):
+    result = {'top': [], 'bottom': [], 'left': [], 'right': []}
+    for i in range(row - 1, -1, -1):
+        result['top'].append(board[i][col])
+    for i in range(row + 1, len(board)):
+        result['bottom'].append(board[i][col])
+    for i in range(col - 1, -1, -1):
+        result['left'].append(board[row][i])
+    for i in range(col + 1, len(board[0])):
+        result['right'].append(board[row][i])
+    return result
+
+
+print(neighbors(1, 2, board))