I was trying to understand why do we add new dimention to our array?
x = np.expand_dims(x, axis=0)
This insert a new axis to array of x. What is the purpose of it?
An example of when expand_dims can be needed:
We have 2 arrays:
a1 = np.arange(20).reshape(4,5)
a2 = np.array([10, 20, 30, 40, 50])
Even though a1 is a 2-D array and a2 is a 1-D array,
it is possible to add them:
a1 + a2
yields:
array([[11, 22, 33, 44, 55],
[16, 27, 38, 49, 60],
[21, 32, 43, 54, 65],
[26, 37, 48, 59, 70]])
i.e. consecutive elements of a2 are added to consecutive columns
in a1 (there are 5 elements in a2 and 5 columns in a1).
But if you have another array:
a3 = np.array([10, 20, 30, 40])
and you want to add each element of a3 to each row in a1 and
you attempt to run:
a1 + a3
you will get ValueError exception.
To successfully perform this operation, a3 must be a columnar array,
i.e.:
it must have 2 dimensions,
it must have the same number of rows as a1,
and a single column.
So, you can run:
a1 + np.expand_dims(a3, axis=1)
getting:
array([[11, 12, 13, 14, 15],
[26, 27, 28, 29, 30],
[41, 42, 43, 44, 45],
[56, 57, 58, 59, 60]])
Related
I have a dataset like:
a1 = [81, 42, 73, 94, 85, 66]
a2 = [63, 55, 79, 65, 94, 76]
a3 = [3, 5, 4, 8, 7, 6]
I want to draw a scatter plot that the x_ticks will be 'a1', 'a2', 'a3'
and the each y_tick is the data in a1, a2 and a3.
for example above a1 x_tick there's 6 dots.[81, 42, 73, 94, 85, 66]
EDIT: Sorry it was a stupid question, and my words aren't explicit as well, I was just trying to draw a simple box plot.
From what I could make of the question, this could be one simple implementation:
import matplotlib.pyplot as plt
a1 = [81, 42, 73, 94, 85, 66]
a2 = [63, 55, 79, 65, 94, 76]
a3 = [3, 5, 4, 8, 7, 6]
plt.scatter(x=[0]*len(a1), y=a1)
plt.scatter(x=[1]*len(a2), y=a2)
plt.scatter(x=[2]*len(a3), y=a3)
plt.xticks(ticks=[0,1,2], labels=["a1", "a2", "a3"])
plt.show()
With the following output:
If you also want to only display the y values at the y axis you can add this line:
plt.yticks(ticks=a1+a2+a3, labels=a1+a2+a3)
But for y values that are very close to each other (see values for a3) this will get crowded.
I am performing an image segmentation with a u-net model.
My mask has classes from 0-50.
I also have a text file dictionary with codes representing each class.
For example -
{1: '1234', 2:'5678', 3:'1245'} etc.
How do I combine when the 2 first string characters are the same so for example above key 1 and 3 are the same because they both start with "12".
How can I do this for all classes?
firstTwoCharDict = {}
for key, value in dictionary.items():
if key == 0:
value == value
firstTwoCharDict[key] = value
else:
value = value[:2]
firstTwoCharDict[key] = value
newDict = {}
for key, value in firstTwoCharDict.items():
if value not in newDict:
newDict[value] = [key]
else:
newDict[value].append(key)
This provides this
{'62': [1, 39],
'90': [2, 5, 9, 20, 32, 42, 47, 72, 88, 91, 95],
'97': [3, 49, 55],
'98': [4, 24, 34, 40, 53, 76, 81, 90, 96],
'31': [6, 17, 30, 48, 83],
'69': [7, 13, 15, 16, 27, 44, 51, 54, 56, 75],
'79': [8, 50],
'71': [10, 19, 22, 35, 61, 63, 65],
'99': [11, 12, 21, 46, 52, 69, 78, 84, 89],
'48': [14, 36, 74],
'60': [18],
'64': [23, 38, 66, 97]
```
Now i have an 2d array with integers, how do I replace them with they keys if the array values are equal to the values in the dict?
I'm trying to compare 2 dataframes and highlight the differences in the second one like this:
I have tried using concat and drop duplicates but I am not sure how to check for the specific cells and also how to highlight them at the end
Possible solution is the following:
import pandas as pd
# set test data
data1 = {"A": [10, 11, 23, 44], "B": [22, 23, 56, 55], "C": [31, 21, 34, 66], "D": [25, 45, 21, 45]}
data2 = {"A": [10, 11, 23, 44, 56, 23], "B": [44, 223, 56, 55, 73, 56], "C": [31, 21, 45, 66, 22, 22], "D": [25, 45, 26, 45, 34, 12]}
# create dataframes
df1 = pd.DataFrame(data1)
df2 = pd.DataFrame(data2)
# define function to highlight differences in dataframes
def highlight_diff(data, other, color='yellow'):
attr = 'background-color: {}'.format(color)
return pd.DataFrame(np.where(data.ne(other), attr, ''),
index=data.index, columns=data.columns)
# apply style using function
df2.style.apply(highlight_diff, axis=None, other=df1)
Returns
I have a small matrix A with dimensions MxNxO
I have a large matrix B with dimensions KxMxNxP, with P>O
I have a vector ind of indices of dimension Ox1
I want to do:
B[1,:,:,ind] = A
But, the lefthand of my equation
B[1,:,:,ind].shape
is of dimension Ox1xMxN and therefore I can not broadcast A (MxNxO) into it.
Why does accessing B in this way change the dimensions of the left side?
How can I easily achieve my goal?
Thanks
There's a feature, if not a bug, that when slices are mixed in the middle of advanced indexing, the sliced dimensions are put at the end.
Thus for example:
In [204]: B = np.zeros((2,3,4,5),int)
In [205]: ind=[0,1,2,3,4]
In [206]: B[1,:,:,ind].shape
Out[206]: (5, 3, 4)
The 3,4 dimensions have been placed after the ind, 5.
We can get around that by indexing first with 1, and then the rest:
In [207]: B[1][:,:,ind].shape
Out[207]: (3, 4, 5)
In [208]: B[1][:,:,ind] = np.arange(3*4*5).reshape(3,4,5)
In [209]: B[1]
Out[209]:
array([[[ 0, 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]]])
This only works when that first index is a scalar. If it too were a list (or array), we'd get an intermediate copy, and couldn't set the value like this.
https://docs.scipy.org/doc/numpy-1.15.0/reference/arrays.indexing.html#combining-advanced-and-basic-indexing
It's come up in other SO questions, though not recently.
weird result when using both slice indexing and boolean indexing on a 3d array
i am newbie and just want to implement Hierarchical Agglomerative clustering for RGB images. For this I extract all values of RGB from an image. And I process image.Next I find its distance and then develop the linkage. Now from linkage I want to extract my original data (i.e RGB values) on specified indices with indices id. Here is code I have done so far.
image = Image.open('image.jpg')
image = image.convert('RGB')
im = np.array(image).reshape((-1,3))
rgb = list(im.getdata())
X = pdist(im)
Y = linkage(X)
I = inconsistent(Y)
based on the 4th column of consistency. I opt minimum value of the cutoff in order to get maximum clusters.
cutoff = 0.7
cluster_assignments = fclusterdata(Y, cutoff)
# Print the indices of the data points in each cluster.
num_clusters = cluster_assignments.max()
print "%d clusters" % num_clusters
indices = cluster_indices(cluster_assignments)
ind = np.array(enumerate(rgb))
for k, ind in enumerate(indices):
print "cluster", k + 1, "is", ind
dendrogram(Y)
I got results like this
cluster 6 is [ 6 11]
cluster 7 is [ 9 12]
cluster 8 is [15]
Means cluster 6 contains the indices of 6 and 11 leafs. Now at this point I stuck in how to map these indices to get original data(i.e rgb values). indices of each rgb values to each pixel in the image. And then I have to generate codebook to implement Agglomeration Clustering. I have no idea how to approach this task. Read a lot of stuff but nothing clued.
Here is my solution:
import numpy as np
from scipy.cluster import hierarchy
im = np.array([[54,101,9],[ 67,89,27],[ 67,85,25],[ 55,106,1],[ 52,108,0],
[ 55,78,24],[ 19,57,8],[ 19,46,0],[ 95,110,15],[112,159,57],
[ 67,118,26],[ 76,127,35],[ 74,128,30],[ 25,62,0],[100,120,9],
[127,145,61],[ 48,112,25],[198,25,21],[203,11,10],[127,171,60],
[124,173,45],[120,133,19],[109,137,18],[ 60,85,0],[ 37,0,0],
[187,47,20],[127,170,52],[ 30,56,0]])
groups = hierarchy.fclusterdata(im, 0.7)
idx_sorted = np.argsort(groups)
group_sorted = groups[idx_sorted]
im_sorted = im[idx_sorted]
split_idx = np.where(np.diff(group_sorted) != 0)[0] + 1
np.split(im_sorted, split_idx)
output:
[array([[203, 11, 10],
[198, 25, 21]]),
array([[187, 47, 20]]),
array([[127, 171, 60],
[127, 170, 52]]),
array([[124, 173, 45]]),
array([[112, 159, 57]]),
array([[127, 145, 61]]),
array([[25, 62, 0],
[30, 56, 0]]),
array([[19, 57, 8]]),
array([[19, 46, 0]]),
array([[109, 137, 18],
[120, 133, 19]]),
array([[100, 120, 9],
[ 95, 110, 15]]),
array([[67, 89, 27],
[67, 85, 25]]),
array([[55, 78, 24]]),
array([[ 52, 108, 0],
[ 55, 106, 1]]),
array([[ 54, 101, 9]]),
array([[60, 85, 0]]),
array([[ 74, 128, 30],
[ 76, 127, 35]]),
array([[ 67, 118, 26]]),
array([[ 48, 112, 25]]),
array([[37, 0, 0]])]