Flatten only part of a dataframe shape for Euclidean calculation? - python

I have a data frame with shape:
(20,30,1024)
I want to find the Euclidean distance between every entry and every other entry in the dataframe (ideally non-redundantly, i.e. don't find the distance of row 1 and 5....and then row 5 and 1 but not there yet). I have this code:
from scipy.spatial.distance import pdist,squareform
distances = pdist(df_test,metric='euclidean')
dist_matrix = squareform(distances)
print(dist_matrix)
The error says:
A 2-dimensional array must be passed.
So I guess I want to convert my matrix from shape (20,30,1024) to (20,30720), and then calculate the pdist/squareform between the rows (i.e. 20 rows of vectors that are 30720 in length).
I know that I can use test_df[0:20].flatten().tolist()
But that completely flattened my matrix, the output shape was (1,614400).
Can someone show me how to convert a shape from (20,30,1024) to (20,3072), or if i'm not going about this the right way?
The ultimate end goal is to calculate Euclidean distance between all non-redundant pairs in a data set, but the data set is big, so I need to do it as efficiently as possible/not duplicating calculations.

The most straightforward way to reshape that I can think of, according to how you described the problem, is:
df_test.values.reshape(20, -1)
By calling .values, you are retrieving your dataframe data as a numpy array. From there, .reshape finishes your job. Since you need a 2D-array, you provide the size of the first dimension (in your case, 20), and by passing -1 Numpy will calculate the size of the second dimension for you (in this case it will multiply the remaining dimension sizes in the original 3D-array)

Related

Subtract the same scalar from all elements in a sparse.csr_matrix

Say I have a csr-matrix,cossim, representing the cosine-similarity of some data. Since I need a distance metric I can create that simply by 1-cossim or -(cossim-1)i.e substract the value of each element from 1, elementwise.
The issue is that it is not implemented thus I thought of doing it by creating a matrix with 1 all over, ones, and then calculate distance = ones-cossim (but I cannot figure out on how to create a "ones matrix" with the same shape, since csr_matrix(np.ones(cossim.shape)) needs 600GB of memory)
Can't we do some elementwise substraction with a sparse matrix?
If not what is the best way to do so?

Efficiently filter 3D matrix in numpy with variable 2D masks

I have a 3D numpy array points of dimensions [10000x3000x128] where the first dimension is the number of frames, the second dimension the number of points in each frame and the third dimension is a 128-element feature vector associated to each point. What I want to do is to efficiently filter the points in each frame by using a boolean 2D mask of dimensions [10000x3000] and for each of the selected points also take the related 128-dim vector of features. Moreover, in output I need still a 3D vector and not a merged 2D vector and possibly avoid any for loop.
Actually what I'm doing is:
# example of points
points = np.array([10000, 3000, 128])
# fg, bg = 2D dimensional boolean np.array
# init empty lists
fg_points, bg_points = [], []
for i in range(points.shape[0]):
fg_mask_tmp, bg_mask_tmp = fg[i], bg[i]
fg_points.append(points[i,fg_mask_tmp,:])
bg_points.append(points[i,bg_mask_tmp,:])
fg_features, bg_features = np.array(fg_points), np.array(bg_points)
But this is a quite naive solution that for sure can be improved in a more numpy-like way.
In addition, I also tried other solutions as:
fg_features = points[fg,:]
But this solution does not preserve the dimensions of the array merging the two first dimensions since the number of filtered points for each frame can vary.
Another solution I tried is to enlarge the 2D masks by appending a [128] true value to the last dimension, but with any successful result.
Dos anyone know a possible efficient solution?
Thank you in advance for any help!

How important are the rows vs columns in PCA?

So i have a dataset with pictures, where each column consist of a vector that can be reshaped into a 32x32 picture. The specific dimensions of my dataset is the following 1024 x 20000. Meaning 20000 samples of images.
Now when i look at various ways of doing PCA without using the built in functions from something like scikit-learn people tend to take either the mean of the rows and subtract the resulting matrix from the original one to get the covariance matrix. I.e the following
A = (1024x20000) #dimensions of the numpy array
mean_rows = A.mean(0)
new_A = A-mean_rows
Other times people tend to get the mean of the columns and the subtract that from the original matrix.
A = (1024x20000) #dimensions of the numpy array
mean_rows = A.mean(1)
new_A = A-mean_rows
Now my question is, when are you supposed to do what? Say i have a dataset as my example which of the methods would i use?
Looked at a variety of websites such as https://machinelearningmastery.com/calculate-principal-component-analysis-scratch-python/,
http://sebastianraschka.com/Articles/2014_pca_step_by_step.html
I think you're talking about normalizing the dataset to have zero mean. You should compute the mean across the axis that contains each observation.
In your example, you have 20,000 observations with 1,024 dimensions each and your matrix has laid out each observation as a column so you should compute the mean of the columns.
In code that would be:
A = A - A.mean(axis=0)

Cosine similarity between two ndarrays

I have two numpy arrays, first array is of size 100*4*200, and second array is of size 150*6*200. In fact, I am storing the 100 samples of 200 dimensional vector representations of 4 fields in array 1 and 140 samples of 200 dimensional vectors of 6 fields in array 2.
Now I want to compute the similarity vector between the samples and create a similarity matrix. For each sample, I would like to calculate the similarity between the each combination of fields and store it such that I get a 15000*24 dimensional array.
First 150 rows will be the similarity vector between 1st row of array 1 and 150 rows of array 2, next 150 rows will be the similarity vector between the 2nd row of array 1 and 150 rows of array 2 etc.
Each similarity vector is # fields in array 1 * # fields in array 2 i.e. 1st element of the similarity vector is cosine similarity between field 1 of array 1 and field 1 of array 2, 2nd element will be the similarity between field 1 of array 1 and field 2 of array 2 and so on with last element is the similarity between last field of array 1 and last field of array 2.
What is the best way to do this using numpy arrays ?
So every "row" (i assume the first axis, that I'll call axis 0) is the sample axis. That means you have 100 samples from one vector, each with fieldsxdimentions 4x200.
Doing this the way you describe, then the first row of the first array would have (4,200) and the second one would then have (150,6,200). Then you'd want to do a cos distance between an (m,n), and (m,n,k) array, which does not make sense (the closest you have to a dot product here would be the tensor product, which I'm fairly sure is not what you want).
So we have to extract these first and then iterate over all the others.
To do this I actually recomend just splitting the array with np.split and iterate over both of them. This is just because I've never come across a faster way in numpy. You could use tensorflow to gain efficiency, but I'm not going into that here in my answer.
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
a = np.random.rand(100, 4, 200)
b = np.random.rand(150, 6, 200)
# We know the output will be 150*100 x 6*4
c = np.empty([15000, 24])
# Make an array with the rows of a and same for b
a_splitted=np.split(a, a.shape[0], 0)
b_splitted=np.split(b, b.shape[0], 0)
i=0
for alpha in a_splitted:
for beta in b_splitted:
# Gives a 4x6 matrix
sim=cosine_similarity(alpha[0],beta[0])
c[i,:]=sim.ravel()
i+=1
For the similarity-function above I just chose what #StefanFalk sugested: sklearn.metrics.pairwise.cosine_similarity. If this similarity measure is not sufficient, then you could either write your own.
I am not at all claiming that this is the best way to do this in all of python. I think the most efficient way is to do this symbolically using, as mentioned, tensorflow.
Anyways, hope it helps!

Auto broadcasting in Scipy

I have two np.ndarrays, data with shape (8000, 500) and sample with shape (1, 500).
What I am trying to achieve is measure various types of metrics between every row in data to sample.
When using from sklearn.metrics.pairwise.cosine_distances I was able to take advantage of numpy's broadcasting executing the following line
x = cosine_distances(data, sample)
But when I tried to use the same procedure with scipy.spatial.distance.cosine I got the error
ValueError: Input vector should be 1-D.
I guess this is a broadcasting issue and I'm trying to find a way to get around it.
My ultimate goal is to iterate over all of the distances available in scipy.spatial.distance that can accept two vectors and apply them to the data and the sample.
How can I replicate the broadcasting that automatically happens in sklearn's in my scipy version of the code?
OK, looking at the docs, http://scikit-learn.org/stable/modules/generated/sklearn.metrics.pairwise.cosine_distances.html
With (800,500) and (1,500) inputs ((samples, features)), you should get back a (800,1) result ((samples1, samples2)).
I wouldn't describe that as broadcasting. It's more like dot product, that performs some sort calculation (norm) over features (the 500 shape), reducing that down to one value. It's more like np.dot(data, sample.T) in its handling of dimensions.
https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.distance.cosine.html is Computes the Cosine distance between 1-D arrays, more like
for row in data:
for s in sample:
d = cosine(row, s)
or since sample has only one row
distances = np.array([cosine(row, sample[0]) for row in data])
In other words, the sklearn version does the pairwise iteration (maybe in compiled code), while the spartial just evaluates the distance for one pair.
pairwise.cosine_similarity does
# K(X, Y) = <X, Y> / (||X||*||Y||)
K = safe_sparse_dot(X_normalized, Y_normalized.T, dense_output=dense_output)
That's the dot like behavior that I mentioned earlier, but with the normalization added.

Categories