Outer product, vectorial operation and loop - numpy - python

I have two arrays of size 15 : A = [a_0, ... , a_14] and B = [b_0, ..., b_14]
Goal: obtain the array C of size 8 resulting from
C = [a_0] * [b_7, ..., b_14] + [a_2, a_3] * [b_3, b_4, b_5, b_6] + [a_3, a_4, a_5, a_6] * [b_2, b_3] + [a_7, ..., a_14] * [b_0]
where * is the outer product np.outer. Note that:
each sub-array is of length 2^i for i between 0 and 3.
from the outer product, we obtain two vectors of size (8) and two matrices of sizes (2, 4) and (4, 2). We suppose that we flatten immediately after the product, in order to be able to sum the four products and have at the end a long vector of size 8.
My implementation is the following:
inds = [0, 1, 3, 7, 15]
C = np.zeros(8)
d = 4
for i in range(d):
left = A[inds[i]:inds[i+1]]
right = B[inds[d-i-1]:inds[d-i]]
C += (left[:, None]*right[None, :]).ravel() # same as np.outer(left, right).ravel()
Question: what is the fastest way to obtain C ? i.e. is there a way to avoid having this for loop to perform the summation ?
If not: what are my options ? code in C++ ? Cython ?
NB: this is to be generalized for loops of range(L+1) with L any integer. In the example above I have illustrated the case L=3 for better comprehension. FYI, the generalized code would look like this:
L = 3
inds = np.cumsum([2**k for k in range(0, L+1)])
inds = np.concatenate(([0], inds))
# Input arrays A and B are of size inds[-1]
C = np.zeros(2**L)
d = L+1
for i in range(d):
left = A[inds[i]:inds[i+1]]
right = B[inds[d-i-1]:inds[d-i]]
C += (left[:, None]*right[None, :]).ravel() # same as np.outer(left, right).ravel()

I think you can simply do:
C = np.outer(A[0], B[7:])+\
np.outer(A[[2,3]], B[[3,4,5,6]]).ravel()+\
np.outer(A[[3,4,5,6]], B[[2,3]]).ravel()+\
np.outer(A[7:], B[0]).ravel()
Am I wrong?

Related

Taking equal number of elements from two arrays, such that the taken values have as few duplicates as possible

Consider we have 2 arrays of size N, with their values in the range [0, N-1]. For example:
a = np.array([0, 1, 2, 0])
b = np.array([2, 0, 3, 3])
I need to produce a new array c which contains exactly N/2 elements from a and b respectively, i.e. the values must be taken evenly/equally from both parent arrays.
(For odd length, this would be (N-1)/2 and (N+1)/2. Can also ignore odd length case, not important).
Taking equal number of elements from two arrays is pretty trivial, but there is an additional constraint: c should have as many unique numbers as possible / as few duplicates as possible.
For example, a solution to a and b above is:
c = np.array([b[0], a[1], b[2], a[3]])
>>> c
array([2, 1, 3, 0])
Note that the position/order is preserved. Each element of a and b that we took to form c is in same position. If element i in c is from a, c[i] == a[i], same for b.
A straightforward solution for this is simply a sort of path traversal, easy enough to implement recursively:
def traverse(i, a, b, path, n_a, n_b, best, best_path):
if n_a == 0 and n_b == 0:
score = len(set(path))
return (score, path.copy()) if score > best else (best, best_path)
if n_a > 0:
path.append(a[i])
best, best_path = traverse(i + 1, a, b, path, n_a - 1, n_b, best, best_path)
path.pop()
if n_b > 0:
path.append(b[i])
best, best_path = traverse(i + 1, a, b, path, n_a, n_b - 1, best, best_path)
path.pop()
return best, best_path
Here n_a and n_b are how many values we will take from a and b respectively, it's 2 and 2 as we want to evenly take 4 items.
>>> score, best_path = traverse(0, a, b, [], 2, 2, 0, None)
>>> score, best_path
(4, [2, 1, 3, 0])
Is there a way to implement the above in a more vectorized/efficient manner, possibly through numpy?
The algorithm is slow mainly because it runs in an exponential time. There is no straightforward way to vectorize this algorithm using only Numpy because of the recursion. Even if it would be possible, the huge number of combinations would cause most Numpy implementations to be inefficient (due to large Numpy arrays to compute). Additionally, there is AFAIK no vectorized operation to count the number of unique values of many rows efficiently (the usual way is to use np.unique which is not efficient in this case and cannot be use without a loop). As a result, there is two possible strategy to speed this up:
trying to find an algorithm with a reasonable complexity (eg. <= O(n^4));
using compilation methods, micro-optimizations and tricks to write a faster brute-force implementation.
Since finding a correct sub-exponential algorithm turns out not to be easy, I choose the other approach (though the first approach is the best).
The idea is to:
remove the recursion by generating all possible solutions using a loop iterating on integer;
write a fast way to count unique items of an array;
use the Numba JIT compiler so to optimize the code that is only efficient once compiled.
Here is the final code:
import numpy as np
import numba as nb
# Naive way to count unique items.
# This is a slow fallback implementation.
#nb.njit
def naive_count_unique(arr):
count = 0
for i in range(len(arr)):
val = arr[i]
found = False
for j in range(i):
if arr[j] == val:
found = True
break
if not found:
count += 1
return count
# Optimized way to count unique items on small arrays.
# Count items 2 by 2.
# Fast on small arrays.
#nb.njit
def optim_count_unique(arr):
count = 0
for i in range(0, len(arr), 2):
if arr[i] == arr[i+1]:
tmp = 1
for j in range(i):
if arr[j] == arr[i]: tmp = 0
count += tmp
else:
val1, val2 = arr[i], arr[i+1]
tmp1, tmp2 = 1, 1
for j in range(i):
val = arr[j]
if val == val1: tmp1 = 0
if val == val2: tmp2 = 0
count += tmp1 + tmp2
return count
#nb.njit
def count_unique(arr):
if len(arr) % 2 == 0:
return optim_count_unique(arr)
else:
# Odd case: not optimized yet
return naive_count_unique(arr)
# Count the number of bits in a 32-bit integer
# See https://stackoverflow.com/questions/71097470/msb-lsb-popcount-in-numba
#nb.njit('int_(uint32)', inline='always')
def popcount(v):
v = v - ((v >> 1) & 0x55555555)
v = (v & 0x33333333) + ((v >> 2) & 0x33333333)
c = np.uint32((v + (v >> 4) & 0xF0F0F0F) * 0x1010101) >> 24
return c
# Count the number of bits in a 64-bit integer
#nb.njit(inline='always')
def bit_count(n):
if n < (1 << 30):
return popcount(np.uint32(n))
else:
return popcount(np.uint32(n)) + popcount(np.uint32(n >> 32))
# Mutate `out` so not to create an expensive new temporary array
#nb.njit
def int_to_path(n, out, a, b):
for i in range(len(out)):
out[i] = a[i] if ((n >> i) & 1) else b[i]
#nb.njit(['(int32[:], int32[:], int64, int64)', '(int64[:], int64[:], int64, int64)'])
def traverse_fast(a, b, n_a, n_b):
# This assertion is needed because the paths are encoded using 64-bit.
# This should not be a problem in practice since the number of solutions to
# test would be impracticably huge to test using this algorithm anyway.
assert n_a + n_b < 62
max_iter = 1 << (n_a + n_b)
path = np.empty(n_a + n_b, dtype=a.dtype)
score, best_score, best_i = 0, 0, 0
# Iterate over all cases (more than the set of possible solution)
for i in range(max_iter):
# Filter the possible solutions
if bit_count(i) != n_b:
continue
# Analyse the score of the solution
int_to_path(i, path, a, b)
score = count_unique(path)
# Store it if it better than the previous one
if score > best_score:
best_score = score
best_i = i
int_to_path(best_i, path, a, b)
return best_score, path
This implementation is about 30 times faster on arrays of size 8 on my machine. On could use several cores to speed this up even further. However, I think it is better to focus on finding a sub-exponential implementation so to avoid wasting more computing resources. Note that the path is different from the initial function but the score is the same on random arrays. It can help others to test their implementation on larger arrays without waiting for a long time.
Test this heavily.
import numpy as np
from numpy.random._generator import default_rng
rand = default_rng(seed=1)
n = 16
a = rand.integers(low=0, high=n, size=n)
b = rand.integers(low=0, high=n, size=n)
uniques = np.setxor1d(a, b)
print(a)
print(b)
print(uniques)
def limited_uniques(arr: np.ndarray) -> np.ndarray:
choose = np.zeros(shape=n, dtype=bool)
_, idx, _ = np.intersect1d(arr, uniques, return_indices=True)
idx = idx[:n//2]
choose[idx] = True
n_missing = n//2 - len(idx)
counts = choose.cumsum()
diffs = np.arange(n) - counts
at = np.searchsorted(diffs, n_missing)
choose[:at] = True
return arr[choose]
a_half = limited_uniques(a)
uniques = np.union1d(uniques, np.setdiff1d(a, a_half))
interleaved = np.empty_like(a)
interleaved[0::2] = a_half
interleaved[1::2] = limited_uniques(b)
print(interleaved)
[ 7 8 12 15 0 2 13 15 3 4 13 6 4 13 4 6]
[10 8 1 0 13 12 13 8 13 5 7 12 1 4 1 7]
[ 1 2 3 5 6 10 15]
[ 7 10 8 8 12 1 15 0 0 13 2 12 3 5 6 4]

How to use numpy to generate random numbers on segmentation intervals

I am using numpy module in python to generate random numbers. When I need to generate random numbers in a continuous interval such as [a,b], I will use
(b-a)*np.random.rand(1)+a
but now I Need to generate a uniform random number in the interval [a, b] and [c, d], what should I do?
I want to generate a random number that is uniform over the length of all the intervals. I do not select an interval with equal probability, and then generate a random number inside the interval. If [a, b] and [c, d] are equal in length, There is no problem with this use, but when the lengths of the intervals are not equal, the random numbers generated by this method are not completely uniform.
You could do something like
a,b,c,d = 1,2,7,9
N = 10
r = np.random.uniform(a-b,d-c,N)
r += np.where(r<0,b,c)
r
# array([7.30557415, 7.42185479, 1.48986144, 7.95916547, 1.30422703,
# 8.79749665, 8.19329762, 8.72669862, 1.88426196, 8.33789181])
You can use
np.random.uniform(a,b)
for your random numbers between a and b (including a but excluding b)
So for random number in [a,b] and [c,d], you can use
np.random.choice( [np.random.uniform(a,b) , np.random.uniform(c,d)] )
Here's a recipe:
def random_multiinterval(*intervals, shape=(1,)):
# FIXME assert intervals are valid and non-overlapping
size = sum(i[1] - i[0] for i in intervals)
v = size * np.random.rand(*shape)
res = np.zeros_like(v)
for i in intervals:
res += (0 < v) * (v < (i[1] - i[0])) * (i[0] + v)
v -= i[1] - i[0]
return res
In [11]: random_multiinterval((1, 2), (3, 4))
Out[11]: array([1.34391171])
In [12]: random_multiinterval((1, 2), (3, 4), shape=(3, 3))
Out[12]:
array([[1.42936024, 3.30961893, 1.01379663],
[3.19310627, 1.05386192, 1.11334538],
[3.2837065 , 1.89239373, 3.35785566]])
Note: This is uniformly distributed over N (non-overlapping) intervals, even if they have different sizes.
You can just assign a probability for how likely it will be [a,b] or [c,d] and then generate accordingly:
import numpy as np
import random
random_roll = random.random()
a = 1
b = 5
c = 7
d = 10
if random_roll > .5: # half the time we will use [a,b]
my_num = (b - a) * np.random.rand(1) + a
else: # the other half we will use [c,d]
my_num = (d - c) * np.random.rand(1) + c
print(my_num)

Numpy subtraction from two arrays

I have two numpy arrays like below
a=np.array([11,12])
b=np.array([9])
#a-b should be [2,12]
I want to subtract both a & b such that result should [2,12]. How can I achieve this result?
You can zero-pad one of the array.
import numpy as np
n = max(len(a), len(b))
a_pad = np.pad(a, (0, n - len(a)), 'constant')
b_pad = np.pad(b, (0, n - len(b)), 'constant')
ans = a_pad - b_pad
Here np.pad's second argument is (#of left pads, #of right pads)
A similar method to #BlownhitherMa, would be to create an array of zeros the size of a (we can call it c), then put in b's values where appropriate:
c = np.zeros_like(a)
c[np.indices(b.shape)] = b
>>> c
array([9, 0])
>>> a-c
array([ 2, 12])
You could use zip_longest from itertools:
import numpy as np
from itertools import zip_longest
a = np.array([11, 12])
b = np.array([9])
result = np.array([ai - bi for ai, bi in zip_longest(a, b, fillvalue=0)])
print(result)
Output
[ 2 12]
Here is a very long laid out solution.
diff =[]
n = min(len(a), len(b))
for i in range (n):
diff.append(a[i] - b[i])
if len(a) > n:
for i in range(n,len(a)):
diff.append(a[i])
elif len(b) > n:
for i in range(n,len(b)):
diff.append(b[i])
diff=np.array(diff)
print(diff)
We can avoid unnecessary padding / temporaries by copying a and then subtracting b in-place:
# let numpy determine appropriate dtype
dtp = (a[:0]-b[:0]).dtype
# copy a
d = a.astype(dtp)
# subtract b
d[:b.size] -= b

Cumulative addition/multiplication in NumPy

Have a relatively simple block of code that loops through two arrays, multiplies, and adds cumulatively:
import numpy as np
a = np.array([1, 2, 4, 6, 7, 8, 9, 11])
b = np.array([0.01, 0.2, 0.03, 0.1, 0.1, 0.6, 0.5, 0.9])
c = []
d = 0
for i, val in enumerate(a):
d += val
c.append(d)
d *= b[i]
Is there a way to do this without iterating? I imagine cumsum/cumprod could be used but I'm having trouble figuring out how. When you break down what's happening step by step, it looks like this:
# 0: 0 + a[0]
# 1: ((0 + a[0]) * b[0]) + a[1]
# 2: ((((0 + a[0]) * b[0]) + a[1]) * b[1]) + a[2]
Edit for clarification: Am interested in the list (or array) c.
In each iteration, you have -
d[n+1] = d[n] + a[n]
d[n+1] = d[n+1] * b[n]
Thus, essentially -
d[n+1] = (d[n] + a[n]) * b[n]
i.e. -
d[n+1] = (d[n]* b[n]) + K[n] #where `K[n] = a[n] * b[n]`
Now, using this formula if you write down the expressions for until n = 2 cases, you would have -
d[1] = d[0]*b[0] + K[0]
d[2] = d[0]*b[0]*b[1] + K[0]*b[1] + K[1]
d[3] = d[0]*b[0]*b[1]*b[2] + K[0]*b[1]*b[2] + K[1]*b[2] + K[2]
Scalars : b[0]*b[1]*b[2] b[1]*b[2] b[2] 1
Coefficients : d[0] K[0] K[1] K[2]
Thus, you would need reversed cumprod of b, perform elementwise multiplication with K array. Finally, to get c, perform cumsum and since c is stored before scaling down by b, so you would need to scale down the cumsum version by the reversed cumprod of b.
The final implementation would look like this -
# Get reversed cumprod of b and pad with `1` at the end
b_rev_cumprod = b[::-1].cumprod()[::-1]
B = np.hstack((b_rev_cumprod,1))
# Get K
K = a*b
# Append with 0 at the start, corresponding starting d
K_ext = np.hstack((0,K))
# Perform elementwsie multiplication and cumsum and scale down for final c
sums = (B*K_ext).cumsum()
c = sums[1:]/b_rev_cumprod
Runtime tests and verify output
Function definitions -
def original_approach(a,b):
c = []
d = 0
for i, val in enumerate(a):
d = d+val
c.append(d)
d = d*b[i]
return c
def vectorized_approach(a,b):
b_rev_cumprod = b[::-1].cumprod()[::-1]
B = np.hstack((b_rev_cumprod,1))
K = a*b
K_ext = np.hstack((0,K))
sums = (B*K_ext).cumsum()
return sums[1:]/b_rev_cumprod
Runtimes and verification
Case #1: OP Sample case
In [301]: # Inputs
...: a = np.array([1, 2, 4, 6, 7, 8, 9, 11])
...: b = np.array([0.01, 0.2, 0.03, 0.1, 0.1, 0.6, 0.5, 0.9])
...:
In [302]: original_approach(a,b)
Out[302]:
[1,
2.0099999999999998,
4.4020000000000001,
6.1320600000000001,
7.6132059999999999,
8.7613205999999995,
14.256792359999999,
18.128396179999999]
In [303]: vectorized_approach(a,b)
Out[303]:
array([ 1. , 2.01 , 4.402 , 6.13206 ,
7.613206 , 8.7613206 , 14.25679236, 18.12839618])
Case #2: Large input case
In [304]: # Inputs
...: N = 1000
...: a = np.random.randint(0,100000,N)
...: b = np.random.rand(N)+0.1
...:
In [305]: np.allclose(original_approach(a,b),vectorized_approach(a,b))
Out[305]: True
In [306]: %timeit original_approach(a,b)
1000 loops, best of 3: 746 µs per loop
In [307]: %timeit vectorized_approach(a,b)
10000 loops, best of 3: 76.9 µs per loop
Please be mindful that for extremely huge input array cases if the b elements are such small fractions, because of cummulative operations, the initial numbers of b_rev_cumprod might come out as zeros resulting in NaNs in those initial places.
Let's see if we can get even faster. I am now leaving the pure python world and show that this purely numeric problems can be optimized even further.
The two players are #Divakar's fast vectorized version:
def vectorized_approach(a,b):
b_rev_cumprod = b[::-1].cumprod()[::-1]
B = np.hstack((b_rev_cumprod,1))
K = a*b
K_ext = np.hstack((0,K))
sums = (B*K_ext).cumsum()
return sums[1:]/b_rev_cumprod
and a cython version:
%%cython
import numpy as np
def cython_approach(long[:] a, double[:] b):
cdef double d
cdef size_t i, n
n = a.shape[0]
cdef double[:] c = np.empty(n)
d = 0
for i in range(n):
d += a[i]
c[i] = d
d *= b[i]
return c
The cython version is about 5x faster than the vectorized version:
%timeit vectorized_approach(a,b) -> 10000 loops, best of 3: 43.4 µs per loop
%timeit cython_approach(a,b) -> 100000 loops, best of 3: 7.7 µs per loop
Another plus of the cython version is that it is much more readable.
The big downside is that you are leaving pure python and depending on your use case compiling an extension module may not be an option for you.
This here works for me and is vectorized
b_mat = np.tile(b,(b.size,1)).T
b_mat = np.vstack((np.ones(b.size),b_mat))
np.fill_diagonal(b_mat,1)
b_mat[np.triu_indices(b.size)]=1
b_prod_mat = np.cumprod(b_mat,axis=0)
b_prod_mat[np.triu_indices(b.size)] = 0
np.fill_diagonal(b_prod_mat,1)
c = np.dot(b_prod_mat,a)
c
# output
array([ 1. , 2.01 , 4.402, 6.132, 7.613, 8.761, 14.257,
18.128, 16.316])
I agree it is not easy to see whats going on. Your array c can be written as a matrix-vector multiplication b_prod_mat * a where a is your array and b_prod_mat consists of specific products of b. All the emphasis is basically to create b_prod_mat.
I am not sure that's better than a for loop but here is a way:
a.dot([np.concatenate((np.zeros(i), (1, ), b[i:-1])) for i in range(len(b))])
What it does it's create line of a big matrix A like this:
1 b0 b0b1 b0b1b2 ... b0b1..bn-1
0 1 b1 b1b2 ... b1..bn-1
0 0 1 b2 ...
...
0 0 0 0 ... 1
Then you simply multiply the vector a with the matrix A and you get your expected result.

Solving Puzzle in Python

I got one puzzle and I want to solve it using Python.
Puzzle:
A merchant has a 40 kg weight which he used in his shop. Once, it fell
from his hands and was broken into 4 pieces. But surprisingly, now he
can weigh any weight between 1 kg to 40 kg with the combination of
these 4 pieces.
So question is, what are weights of those 4 pieces?
Now I wanted to solve this in Python.
The only constraint i got from the puzzle is that sum of 4 pieces is 40. With that I could filter all the set of 4 values whose sum is 40.
import itertools as it
weight = 40
full = range(1,41)
comb = [x for x in it.combinations(full,4) if sum(x)==40]
length of comb = 297
Now I need to check each set of values in comb and try all the combination of operations.
Eg if (a,b,c,d) is the first set of values in comb, I need to check a,b,c,d,a+b,a-b, .................a+b+c-d,a-b+c+d........ and so on.
I tried a lot, but i am stuck at this stage, ie how to check all these combination of calculations to each set of 4 values.
Question :
1) I think i need to get a list all possible combination of [a,b,c,d] and [+,-].
2) does anyone have a better idea and tell me how to go forward from here?
Also, I want to do it completely without help of any external libraries, need to use only standard libraries of python.
EDIT : Sorry for the late info. Its answer is (1,3,9,27), which I found a few years back. I have checked and verified the answer.
EDIT : At present, fraxel's answer works perfect with time = 0.16 ms. A better and faster approach is always welcome.
Regards
ARK
Earlier walk-through anwswer:
We know a*A + b*B + c*C + d*D = x for all x between 0 and 40, and a, b, c, d are confined to -1, 0, 1. Clearly A + B + C + D = 40. The next case is x = 39, so clearly the smallest move is to remove an element (it is the only possible move that could result in successfully balancing against 39):
A + B + C = 39, so D = 1, by neccessity.
next:
A + B + C - D = 38
next:
A + B + D = 37, so C = 3
then:
A + B = 36
then:
A + B - D = 35
A + B - C + D = 34
A + B - C = 33
A + B - C - D = 32
A + C + D = 31, so A = 9
Therefore B = 27
So the weights are 1, 3, 9, 27
Really this can be deduced immediately from the fact that they must all be multiples of 3.
Interesting Update:
So here is some python code to find a minimum set of weights for any dropped weight that will span the space:
def find_weights(W):
weights = []
i = 0
while sum(weights) < W:
weights.append(3 ** i)
i += 1
weights.pop()
weights.append(W - sum(weights))
return weights
print find_weights(40)
#output:
[1, 3, 9, 27]
To further illustrate this explaination, one can consider the problem as the minimum number of weights to span the number space [0, 40]. It is evident that the number of things you can do with each weight is trinary /ternary (add weight, remove weight, put weight on other side). So if we write our (unknown) weights (A, B, C, D) in descending order, our moves can be summarised as:
ABCD: Ternary:
40: ++++ 0000
39: +++0 0001
38: +++- 0002
37: ++0+ 0010
36: ++00 0011
35: ++0- 0012
34: ++-+ 0020
33: ++-0 0021
32: ++-- 0022
31: +0++ 0100
etc.
I have put ternary counting from 0 to 9 alongside, to illustrate that we are effectively in a trinary number system (base 3). Our solution can always be written as:
3**0 + 3**1 +3**2 +...+ 3**N >= Weight
For the minimum N that this holds true. The minimum solution will ALWAYS be of this form.
Furthermore, we can easily solve the problem for large weights and find the minimum number of pieces to span the space:
A man drops a known weight W, it breaks into pieces. His new weights allow him to weigh any weight up to W. How many weights are there, and what are they?
#what if the dropped weight was a million Kg:
print find_weights(1000000)
#output:
[1, 3, 9, 27, 81, 243, 729, 2187, 6561, 19683, 59049, 177147, 531441, 202839]
Try using permutations for a large weight and unknown number of pieces!!
Here is a brute-force itertools solution:
import itertools as it
def merchant_puzzle(weight, pieces):
full = range(1, weight+1)
all_nums = set(full)
comb = [x for x in it.combinations(full, pieces) if sum(x)==weight]
funcs = (lambda x: 0, lambda x: x, lambda x: -x)
for c in comb:
sums = set()
for fmap in it.product(funcs, repeat=pieces):
s = sum(f(x) for x, f in zip(c, fmap))
if s > 0:
sums.add(s)
if sums == all_nums:
return c
>>> merchant_puzzle(40, 4)
(1, 3, 9, 27)
For an explanation of how it works, check out the answer Avaris gave, this is an implementation of the same algorithm.
You are close, very close :).
Since this is a puzzle you want to solve, I'll just give pointers. For this part:
Eg if (a,b,c,d) is the first set of values in comb, i need to check
a,b,c,d,a+b,a-b, .................a+b+c-d,a-b+c+d........ and so on.
Consider this: Each weight can be put to one scale, the other or neither. So for the case of a, this can be represented as [a, -a, 0]. Same with the other three. Now you need all possible pairings with these 3 possibilities for each weight (hint: itertools.product). Then, a possible measuring of a pairing (lets say: (a, -b, c, 0)) is merely the sum of these (a-b+c+0).
All that is left is just checking if you could 'measure' all the required weights. set might come handy here.
PS: As it was stated in the comments, for the general case, it might not be necessary that these divided weights should be distinct (for this problem it is). You might reconsider itertools.combinations.
I brute forced the hell out of the second part.
Do not click this if you don't want to see the answer. Obviously, if I was better at permutations, this would have required a lot less cut/paste search/replace:
http://pastebin.com/4y2bHCVr
I don't know Python syntax, but maybe you can decode this Scala code; start with the 2nd for-loop:
def setTo40 (a: Int, b: Int, c: Int, d: Int) = {
val vec = for (
fa <- List (0, 1, -1);
fb <- List (0, 1, -1);
fc <- List (0, 1, -1);
fd <- List (0, 1, -1);
prod = fa * a + fb * b + fc * c + fd * d;
if (prod > 0)
) yield (prod)
vec.toSet
}
for (a <- (1 to 9);
b <- (a to 14);
c <- (b to 20);
d = 40-(a+b+c)
if (d > 0)) {
if (setTo40 (a, b, c, d).size > 39)
println (a + " " + b + " " + c + " " + d)
}
With weights [2, 5, 15, 18] you can also measure all objects between 1 and 40kg, although some of them will need to be measured indirectly. For example, to measure an object weighting 39kg, you would first compare it with 40kg and the balance would pend to the 40kg side (because 39 < 40), but then if you remove the 2kg weight it would pend to the other side (because 39 > 38) and thus you can conclude the object weights 39kg.
More interestingly, with weights [2, 5, 15, 45] you can measure all objects up to 67kg.
If anyone doesn't want to import a library to import combos/perms, this will generate all possible 4-move strategies...
# generates permutations of repeated values
def permutationsWithRepeats(n, v):
perms = []
value = [0] * n
N = n - 1
i = n - 1
while i > -1:
perms.append(list(value))
if value[N] < v:
value[N] += 1
else:
while (i > -1) and (value[i] == v):
value[i] = 0
i -= 1
if i > -1:
value[i] += 1
i = N
return perms
# generates the all possible permutations of 4 ternary moves
def strategy():
move = ['-', '0', '+']
perms = permutationsWithRepeats(4, 2)
for i in range(len(perms)):
s = ''
for j in range(4):
s += move[perms[i][j]]
print s
# execute
strategy()
My solution as follows:
#!/usr/bin/env python3
weight = 40
parts = 4
part=[0] * parts
def test_solution(p, weight,show_result=False):
cv=[0,0,0,0]
for check_weight in range(1,weight+1):
sum_ok = False
for parts_used in range(2 ** parts):
for options in range(2 ** parts):
for pos in range(parts):
pos_neg = int('{0:0{1}b}'.format(options,parts)[pos]) * 2 - 1
use = int('{0:0{1}b}'.format(parts_used,parts)[pos])
cv[pos] = p[pos] * pos_neg * use
if sum(cv) == check_weight:
if show_result:
print("{} = sum of:{}".format(check_weight, cv))
sum_ok = True
break
if sum_ok:
continue
else:
return False
return True
for part[0] in range(1,weight-parts):
for part[1] in range(part[0]+1, weight - part[0]):
for part[2] in range( part[1] + 1 , weight - sum(part[0:2])):
part[3] = weight - sum(part[0:3])
if test_solution(part,weight):
print(part)
test_solution(part,weight,True)
exit()
It gives you all the solutions for the given weights
More dynamic than my previous answer, so it also works with other numbers. But breaking up into 5 peaces takes some time:
#!/usr/bin/env python3
weight = 121
nr_of_parts = 5
# weight = 40
# nr_of_parts = 4
weight = 13
nr_of_parts = 3
part=[0] * nr_of_parts
def test_solution(p, weight,show_result=False):
cv=[0] * nr_of_parts
for check_weight in range(1,weight+1):
sum_ok = False
for nr_of_parts_used in range(2 ** nr_of_parts):
for options in range(2 ** nr_of_parts):
for pos in range(nr_of_parts):
pos_neg = int('{0:0{1}b}'.format(options,nr_of_parts)[pos]) * 2 - 1
use = int('{0:0{1}b}'.format(nr_of_parts_used,nr_of_parts)[pos])
cv[pos] = p[pos] * pos_neg * use
if sum(cv) == check_weight:
if show_result:
print("{} = sum of:{}".format(check_weight, cv))
sum_ok = True
break
if sum_ok:
continue
else:
return False
return True
def set_parts(part,position, nr_of_parts, weight):
if position == 0:
part[position] = 1
part, valid = set_parts(part,position+1,nr_of_parts,weight)
return part, valid
if position == nr_of_parts - 1:
part[position] = weight - sum(part)
if part[position -1] >= part[position]:
return part, False
return part, True
part[position]=max(part[position-1]+1,part[position])
part, valid = set_parts(part, position + 1, nr_of_parts, weight)
if not valid:
part[position]=max(part[position-1]+1,part[position]+1)
part=part[0:position+1] + [0] * (nr_of_parts - position - 1)
part, valid = set_parts(part, position + 1, nr_of_parts, weight)
return part, valid
while True:
part, valid = set_parts(part, 0, nr_of_parts, weight)
if not valid:
print(part)
print ('No solution posible')
exit()
if test_solution(part,weight):
print(part,' ')
test_solution(part,weight,True)
exit()
else:
print(part,' ', end='\r')

Categories

Resources