在数组中查找特定值

Question

I'm trying to insert conditioned values in an array into an empty array in exact position.

empty_array = np.zeros([40,100])
for x in range (1,24):
    array[x,:,:] #which is also sized 40x100
    if the_values_in_the_array < 0.25:
         the_values_in_the_array = 0
    empty_array = empty_array + array [x,:,:]

Which syntax should I use for this logic? And how should I scan the_values_in_the_array to find the conditioned values?

Answer 1

empty_array = np.zeros([40,100])
array = np.random.rand(24,40,100)

array[array<0.25]=0 # change all the values which is <0.25 to 0
for x in range(1,24):
    empty_array += array[x,:,:]

Answer 2

I think the is the operation you are trying to do. I suggest using the np.where routine for setting values less than 0.25 to zero. Then you can sum over just the first dimension of the array to get the output array you are looking for. I reduced the dimensions of the problem for the example.

import numpy as np

vals = np.random.random([24, 2, 3])
vals_filtered = np.where(vals < 0.25, 0.0, vals)
out = vals_filtered.sum(axis=0)
print("First sample array has the slice vals[0,:,:]:\n{}\n".format(vals[0, :, :]))
print("First sample array with vals>0.25 set to 0.0:\n{}\n".format(vals_filtered[0, :, :]))
print("Output array is the sum over the first dimension:\n{}\n".format(out))

This returns the following output.

First sample array has the slice vals[0, :, :]: 
[[ 0.16272567  0.13695067  0.5954866 ]
 [ 0.50367823  0.8519252   0.3000367 ]]

First sample array with vals>0.25 set to 0.0: 
[[ 0.          0.          0.5954866 ]
 [ 0.50367823  0.8519252   0.3000367 ]]

Output array is the sum over the first dimension: 
[[ 11.12707813  12.04175706  11.5812803 ]
 [ 13.73036272   9.3988165   12.41808745]]

Is this the calculation you were looking for? Calling vals.sum(axis=0) is a faster way to do the operation. It is usually better to call numpy's built-in array routines, as opposed to an explicit for loop.