In essence this is what I want to create
import numpy as np
N = 100 # POPULATION SIZE
D = 30 # DIMENSIONALITY
lowerB = [-5.12] * D # LOWER BOUND (IN ALL DIMENSIONS)
upperB = [5.12] * D # UPPER BOUND (IN ALL DIMENSIONS)
# INITIALISATION PHASE
X = np.empty([N, D]) # EMPTY FLIES ARRAY OF SIZE: (N,D)
# INITIALISE FLIES WITHIN BOUNDS
for i in range(N):
for d in range(D):
X[i, d] = np.random.uniform(lowerB[d], upperB[d])
but I want to do so without the for loops to save time and use List comprehensions
I have try things like
np.array([(x,y)for x in range(N)for y in range(D)])
but this doesn’t get me to an array like array([100,30]). Does anyone know a tutorial or the correct documentation I should be looking at so I can learn exactly how to do this?
Related
I am trying to manually convert a BGR image to HSV. I need to find the maximum pixel value each of 3 image channels (numPy arrays) and create a new array which contains the maximum of the 3 channels.
def convertBGRtoHSV(image):
# normalize image
scaledImage = image // 256
# split image into 3 channels
B, G, R = cv2.split(scaledImage)
# find the shape of each array
heightB, widthB = B.shape
V = []
for h_i in range(0, height):
for w_i in range(0, width):
V[h_i][w_i] = max(B[h_i][w_i], G[h_i][w_i], R[h_i][w_i])
I am getting this error: IndexError: list index out of range
I know this loop is incorrect. I know to access the value of a pixel in an array you must say the location as such as x[:,:] but I am not sure how to loop over all the pixels of each image and make a new array with the max value of each array element.
If possible I would like to know how to use a numPy "Vectorized Operation" to accomplish this as well as the for loop.
There is a builtin function for element-wise maximum:
V = np.maximum(np.maximum(R, G), B)
... and you are done
Following up on my comment:
import cv2
import numpy as np
image = cv2.imread(image)
height, width, _ = image.shape
# initialize your output array 'v'
v = np.zeros((height, width))
# loop over each index in ranges dictated by the image shape
for row in range(height):
for col in range(width):
# assign the maximum value across the 3rd dimension (color channel)
# from the original image to your output array
v[row, col] = max(image[row, col, :])
My question follows from a previous question that I asked, but without including my own code (which I should have done initially).
Moving rows between subarrays
which solves my dilemma only partially. But I have adopted the method in the below code.
Here is relevant code to my specific problem:
K <- 2 # number of equally-sized (sub)populations
N <- 5 # total number of sampled individuals
Hstar <- 5 # total number of haplotypes
probs <- rep(1/Hstar, Hstar) # haplotype frequencies
m = 0.1 # migration rate between subpopulations
perms <- 10000 # number of permutations
## Set up container(s) to hold the identity of each individual from each permutation ##
num.specs <- ceiling(N / K)
## Create an ID for each haplotype ##
haps <- 1:Hstar
## Assign individuals (N) to each subpopulation (K) ##
specs <- 1:num.specs
## Generate permutations, assume each permutation has N individuals, and sample those individuals' haplotypes from the probabilities ##
gen.perms <- function() {
sample(haps, size = num.specs, replace = TRUE, prob = probs)
}
pop <- array(dim = c(perms, num.specs, K))
for (i in 1:K) {
pop[,, i] <- replicate(perms, gen.perms())
}
## Allow individuals from permutations to migrate between subpopulations ##
for (i in 1:K) {
if (m != 0){
ind <- sample(perms, size = perms * m, replace = FALSE) # sample random row from random subpopulation
}
pop[ind,] ## should swap rows between subarrays, but instead throws an error.
}
'ind' identifies the rows that are to be swapped.
The goal is to swap rows from one subpopulation (= subarray) to the other as initially asked in the linked question. For example, switch row 1 of subarray 1 with row 100 of subarray 2. Most importantly, I need to preserve the array type. In the end, 'pop' must have dimensions = c(perms, num.specs, K). Can this be done?
Any assistance is greatly appreciated.
i think you forget to put a comma. I guess you need to change line 38 to this:
pop[ind,,] ## should swap rows between subarrays, but instead throws an error.
and when you want to go on with the changed pop-array, with swapped lines, you need to store it in a variable. for example like this:
pop_new <- pop[ind,,]
or you can store it in the same variable, which mean it will overwrite the old content of pop. When you use the same vector (ind) for index as well the target variable you replace only the old value. With sample() you swap the rows randomly :
pop[ind,,] <- pop[sample(ind),,]
In this associative lstm paper, http://arxiv.org/abs/1602.03032, they ask to permute a complex tensor.
They have provided their code here: https://github.com/mohammadpz/Associative_LSTM/blob/master/bricks.py#L79
I'm trying to replicate this in tensorflow. Here is what I have done:
# shape: C x F/2
# output = self.permutations: [num_copies x cell_size]
permutations = []
indices = numpy.arange(self._dim / 2) #[1 ,2 ,3 ...64]
for i in range(self._num_copies):
numpy.random.shuffle(indices) #[4, 48, 32, ...64]
permutations.append(numpy.concatenate(
[indices,
[ind + self._dim / 2 for ind in indices]]))
#you're appending a row with two columns -- a permutation in the first column, and the same permutation + dim/2 for imaginary
# C x F (numpy)
self.permutations = tf.constant(numpy.vstack(permutations), dtype = tf.int32) #This is a permutation tensor that has the stored permutations
# output = self.permutations: [num_copies x cell_size]
def permute(complex_tensor): #complex tensor is [batch_size x cell_size]
gather_tensor = tf.gather_nd(complex_tensor, self.permutations)
return gather_tensor
Basically, my question is: How efficiently can this be done in TensorFlow? Is there anyway to keep the batch size dimension fixed of complex tensor?
Also, is gather_nd the best way to go about this? Or is it better to do a for loop and iterate over each row in self.permutations using tf.gather?
def permute(self, complex_tensor):
inputs_permuted = []
for i in range(self.permutations.get_shape()[0].value):
inputs_permuted.append(
tf.gather(complex_tensor, self.permutations[i]))
return tf.concat(0, inputs_permuted)
I thought that gather_nd would be far more efficient.
Nevermind, I figured it out, the trick is to just use permute the original input tensor using tf transpose. This will allow you then to do a tf.gather on the entire matrix. Then you can tf concat the matrices together. Sorry if this wasted anyone's time.
for the following code:
from array import *
x=[]
x.append(0.232)
print (x)
for i in range(25):
x[i+1]=(1/(i+1))-5*x[i]
I have this error:
x[i+1]=(1/(i+1))-5*x[i]
IndexError: list assignment index out of range
This may be happening because I have defined x to be an empty array. But how do I define the array and perform the same operation otherwise?
list is not designed for efficient mathematical operations and therefore its better to use numpy arrays for doing mathematical operations. However, if you want to use list, you may define a list initialized with n zero's using
x=[0]*n
x[0] = 0.232
x[1] = ....
....
Remember, that a multidimensional list created using above approach will refer to same element in the array! For example:
l = [0,0,0]*5
will be creating five same list's inside another list not separate list's. So its a bad idea to create multidimensional array like this!
A better way would be to create arrays using numpy using following code:
from numpy import empty, zeros
x = empty(n) # or # x = zeros(n)
x[0] = 0.232
x[1] = ....
....
and
l = empty((3,5)) # or # l = zeros((3,5))
for a array with 3 rows and 5 columns.
In a for-loop, I run in i over an array which I would like to sub-index in dimension i. How can this be done? So a minimal example would be
(A <- array(1:24, dim = 2:4))
A[2,,] # i=1
A[,1,] # i=2
A[,,3] # i=3
where I index 'by foot'. I tried something along the lines of this but wasn't successful. Of course one could could create "2,," as a string and then eval & parse the code, but that's ugly. Also, inside the for loop (over i), I could use aperm() to permute the array such that the new first dimension is the former ith, so that I can simply access the first component. But that's kind of ugly too and requires to permute the array back. Any ideas how to do it more R-like/elegantly?
The actual problem is for a multi-dimensional table() object, but I think the idea will remain the same.
Update
I accepted Rick's answer. I just present it with a for loop and simplified it further:
subindex <- c(2,1,3) # in the ith dimension, we would like to subindex by subindex[i]
for(i in seq_along(dim(A))) {
args <- list(1:2, 1:3, 1:4)
args[i] <- subindex[i]
print(do.call("[", c(list(A), args)))
}
#Build a multidimensional array
A <- array(1:24, dim = 2:4)
# Select a sub-array
indexNumber = 2
indexSelection = 1
# Build a parameter list indexing all the elements of A
parameters <- list(A, 1:2, 1:3, 1:4)
# Modify the appropriate list element to a single value
parameters[1 + indexNumber] <- indexSelection
# select the desired subarray
do.call("[", parameters)
# Now for something completely different!
#Build a multidimensional array
A <- array(1:24, dim = 2:4)
# Select a sub-array
indexNumber = 2
indexSelection = 1
reduced <- A[slice.index(A, indexNumber) == indexSelection]
dim(reduced) <- dim(A)[-indexNumber]
# Also works on the left-side
A[slice.index(A, 2)==2] <- -1:-8