My code:
from numpy import *
def pca(orig_data):
data = array(orig_data)
data = (data - data.mean(axis=0)) / data.std(axis=0)
u, s, v = linalg.svd(data)
print s #should be s**2 instead!
print v
def load_iris(path):
lines = []
with open(path) as input_file:
lines = input_file.readlines()
data = []
for line in lines:
cur_line = line.rstrip().split(',')
cur_line = cur_line[:-1]
cur_line = [float(elem) for elem in cur_line]
data.append(array(cur_line))
return array(data)
if __name__ == '__main__':
data = load_iris('iris.data')
pca(data)
The iris dataset: http://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data
Output:
[ 20.89551896 11.75513248 4.7013819 1.75816839]
[[ 0.52237162 -0.26335492 0.58125401 0.56561105]
[-0.37231836 -0.92555649 -0.02109478 -0.06541577]
[ 0.72101681 -0.24203288 -0.14089226 -0.6338014 ]
[ 0.26199559 -0.12413481 -0.80115427 0.52354627]]
Desired Output:
Eigenvalues - [2.9108 0.9212 0.1474 0.0206]
Principal Components - Same as I got but transposed so okay I guess
Also, what's with the output of the linalg.eig function? According to the PCA description on wikipedia, I'm supposed to this:
cov_mat = cov(orig_data)
val, vec = linalg.eig(cov_mat)
print val
But it doesn't really match the output in the tutorials I found online. Plus, if I have 4 dimensions, I thought I should have 4 eigenvalues and not 150 like the eig gives me. Am I doing something wrong?
edit: I've noticed that the values differ by 150, which is the number of elements in the dataset. Also, the eigenvalues are supposed to add to be equal to the number of dimensions, in this case, 4. What I don't understand is why this difference is happening. If I simply divided the eigenvalues by len(data) I could get the result I want, but I don't understand why. Either way the proportion of the eigenvalues isn't altered, but they are important to me so I'd like to understand what's going on.
You decomposed the wrong matrix.
Principal Component Analysis requires manipulating the eigenvectors/eigenvalues
of the covariance matrix, not the data itself. The covariance matrix, created from an m x n data matrix, will be an m x m matrix with ones along the main diagonal.
You can indeed use the cov function, but you need further manipulation of your data. It's probably a little easier to use a similar function, corrcoef:
import numpy as NP
import numpy.linalg as LA
# a simulated data set with 8 data points, each point having five features
data = NP.random.randint(0, 10, 40).reshape(8, 5)
# usually a good idea to mean center your data first:
data -= NP.mean(data, axis=0)
# calculate the covariance matrix
C = NP.corrcoef(data, rowvar=0)
# returns an m x m matrix, or here a 5 x 5 matrix)
# now get the eigenvalues/eigenvectors of C:
eval, evec = LA.eig(C)
To get the eigenvectors/eigenvalues, I did not decompose the covariance matrix using SVD,
though, you certainly can. My preference is to calculate them using eig in NumPy's (or SciPy's)
LA module--it is a little easier to work with than svd, the return values are the eigenvectors
and eigenvalues themselves, and nothing else. By contrast, as you know, svd doesn't return these these directly.
Granted the SVD function will decompose any matrix, not just square ones (to which the eig function is limited); however when doing PCA, you'll always have a square matrix to decompose,
regardless of the form that your data is in. This is obvious because the matrix you
are decomposing in PCA is a covariance matrix, which by definition is always square
(i.e., the columns are the individual data points of the original matrix, likewise
for the rows, and each cell is the covariance of those two points, as evidenced
by the ones down the main diagonal--a given data point has perfect covariance with itself).
The left singular values returned by SVD(A) are the eigenvectors of AA^T.
The covariance matrix of a dataset A is : 1/(N-1) * AA^T
Now, when you do PCA by using the SVD, you have to divide each entry in your A matrix by (N-1) so you get the eigenvalues of the covariance with the correct scale.
In your case, N=150 and you haven't done this division, hence the discrepancy.
This is explained in detail here
(Can you ask one question, please? Or at least list your questions separately. Your post reads like a stream of consciousness because you are not asking one single question.)
You probably used cov incorrectly by not transposing the matrix first. If cov_mat is 4-by-4, then eig will produce four eigenvalues and four eigenvectors.
Note how SVD and PCA, while related, are not exactly the same. Let X be a 4-by-150 matrix of observations where each 4-element column is a single observation. Then, the following are equivalent:
a. the left singular vectors of X,
b. the principal components of X,
c. the eigenvectors of X X^T.
Also, the eigenvalues of X X^T are equal to the square of the singular values of X. To see all this, let X have the SVD X = QSV^T, where S is a diagonal matrix of singular values. Then consider the eigendecomposition D = Q^T X X^T Q, where D is a diagonal matrix of eigenvalues. Replace X with its SVD, and see what happens.
Question already adressed: Principal component analysis in Python
Related
I was going through the book called Hands-On Machine Learning with Scikit-Learn, Keras and Tensorflow and the author was explaining how the pseudo-inverse (Moore-Penrose inverse) of a matrix is calculated in the context of Linear Regression. I'm quoting verbatim here:
The pseudoinverse itself is computed using a standard matrix
factorization technique called Singular Value Decomposition (SVD) that
can decompose the training set matrix X into the matrix
multiplication of three matrices U Σ VT (see numpy.linalg.svd()). The
pseudoinverse is calculated as X+ = V * Σ+ * UT. To compute the matrix
Σ+, the algorithm takes Σ and sets to zero all values smaller than a
tiny threshold value, then it replaces all nonzero values with their
inverse, and finally it transposes the resulting matrix. This approach
is more efficient than computing the Normal equation.
I've got an understanding of how the pseudo-inverse and SVD are related from this post. But I'm not able to grasp the rationale behind setting all values less than the threshold to zero. The inverse of a diagonal matrix is obtained by taking the reciprocals of the diagonal elements. Then small values would be converted to large values in the inverse matrix, right? Then why are we removing the large values?
I went and looked into the numpy code, and it looks like follows, just for reference:
#array_function_dispatch(_pinv_dispatcher)
def pinv(a, rcond=1e-15, hermitian=False):
a, wrap = _makearray(a)
rcond = asarray(rcond)
if _is_empty_2d(a):
m, n = a.shape[-2:]
res = empty(a.shape[:-2] + (n, m), dtype=a.dtype)
return wrap(res)
a = a.conjugate()
u, s, vt = svd(a, full_matrices=False, hermitian=hermitian)
# discard small singular values
cutoff = rcond[..., newaxis] * amax(s, axis=-1, keepdims=True)
large = s > cutoff
s = divide(1, s, where=large, out=s)
s[~large] = 0
res = matmul(transpose(vt), multiply(s[..., newaxis], transpose(u)))
return wrap(res)
It's almost certainly an adjustment for numerical error. To see why this might be necessary, look what happens when you take the svd of a rank-one 2x2 matrix. We can create a rank-one matrix by taking the outer product of a vector like so:
>>> a = numpy.arange(2) + 1
>>> A = a[:, None] * a[None, :]
>>> A
array([[1, 2],
[2, 4]])
Although this is a 2x2 matrix, it only has one linearly independent column, and so its rank is one instead of two. So we should expect that when we pass it to svd, one of the singular values will be zero. But look what happens:
>>> U, s, V = numpy.linalg.svd(A)
>>> s
array([5.00000000e+00, 1.98602732e-16])
What we actually get is a singular value that is not quite zero. This result is inevitable in many cases given that we are working with finite-precision floating point numbers. So although the problem you have identified is a real one, we will not be able to tell in practice the difference between a matrix that really has a very small singular value and a matrix that ought to have a zero singular value but doesn't. Setting small values to zero is the safest practical way to handle that problem.
I've got a 2x2 matrix defined by the variables J00, J01, J10, J11 coming in from other inputs. Since the matrix is small, I was able to compute the spectral norm by first computing the trace and determinant
J_T = tf.reduce_sum([J00, J11])
J_ad = tf.reduce_prod([J00, J11])
J_cb = tf.reduce_prod([J01, J10])
J_det = tf.reduce_sum([J_ad, -J_cb])
and then solving the quadratic
L1 = J_T/2.0 + tf.sqrt(J_T**2/4.0 - J_det)
L2 = J_T/2.0 - tf.sqrt(J_T**2/4.0 - J_det)
spectral_norm = tf.maximum(L1, L2)
This works, but it looks rather ugly and it isn't generalizable to larger matrices. Is there cleaner way (maybe a method call that I'm missing) to compute spectral_norm?
The spectral norm of a matrix J equals the largest singular value of the matrix.
Therefore you can use tf.svd() to perform the singular value decomposition, and take the largest singular value:
spectral_norm = tf.svd(J,compute_uv=False)[...,0]
where J is your matrix.
Notes:
I use compute_uv=False since we are interested only in singular values, not singular vectors.
J does not need to be square.
This solution works also for the case where J has any number of batch dimensions (as long as the two last dimensions are the matrix dimensions).
The elipsis ... operation works as in NumPy.
I take the 0 index because we are interested only in the largest singular value.
I am doing a benchmarking test in python on different ways to calculate A'*A, with A being a N x M matrix. One of the fastest ways was to use numpy.dot().
I was curious if I can obtain the same result using numpy.cov() (which gives the covariance matrix) by somehow varying the weights or by somehow pre-processing the A matrix ? But I had no success. Does anyone know if there is any relation between the product A'*A and covariance of A, where A is a matrix with N rows/observations and M columns/variables?
Have a look at the cov source. Near the end of the function it does this:
c = dot(X, X_T.conj())
Which is basically the dot product you want to perform. However, there are all kinds of other operations: checking inputs, subtracting the mean, normalization, ...
In short, np.cov will never ever be faster than np.dot(A.T, A) because internally it contains exactly that operation.
That said - the covariance matrix is computed as
Or in Python:
import numpy as np
a = np.random.rand(10, 3)
m = np.mean(a, axis=0, keepdims=True)
x = np.dot((a - m).T, a - m) / (a.shape[0] - 1)
y = np.cov(a.T)
assert np.allclose(x, y) # check they are equivalent
As you can see, the covariance matrix is equivalent to the raw dot product if you subtract the mean of each variable and divide the result by the number of samples (minus one).
I am trying to understand this optimized code to find cosine similarity between users matrix.
def fast_similarity(ratings,epsilon=1e-9):
# epsilon -> small number for handling dived-by-zero errors
sim = ratings.T.dot(ratings) + epsilon
norms = np.array([np.sqrt(np.diagonal(sim))])
return (sim / norms / norms.T)
If ratings =
items
u [
s [1,2,3]
e [4,5,6]
r [7,8,9]
s ]
nomrs will be equal to = [1^2 + 5^2 + 9^2]
but why we are writing sim/norms/norms.T to calculate cosine similarity?
Any help is appreciated.
Going through the code we have that:
And this means that, one the diagonal of the sim matrix we have the result of the multiplication of each column.
You can give it a try if you want using a simple matrix:
And you can easily check that this gram matrix (that's how this matrix product is named) has this property.
Now the code defines norms that is nothing but an array taking the diagonal of our gram matrix and apply a sqrt on each element of it.
This will give us an array containing the norm value for each column:
So basically the norms vector contains the norm value of each column of the result matrix.
Once we have all those data we can evaluate the cosine similarity between those users, so we know that cosine similarity is evaluated like:
Note that :
So we have that our similarity is going to be:
So we just have to substitute the terms with our code variable to get:
And this explain why you have this line of code:
return sim / norms / norms.T
EDIT:
Since it seems that I was not clear, every time I am talking about matrix multiplication in this answer I am reffering to the DOT PRODUCT of two matrices.
This actually means that when it's written A*B we actually develop and
solve as A.T * B
I'm using the module hcluster to calculate a dendrogram from a distance matrix. My distance matrix is an array of arrays generated like this:
import hcluster
import numpy as np
mols = (..a list of molecules)
distMatrix = np.zeros((10, 10))
for i in range(0,10):
for j in range(0,10):
sim = OETanimoto(mols[i],mols[j]) # a function to calculate similarity between molecules
distMatrix[i][j] = 1 - sim
I then use the command distVec = hcluster.squareform(distMatrix) to convert the matrix into a condensed vector and calculate the linkage matrix with vecLink = hcluster.linkage(distVec).
All this works fine but if I calculate the linkage matrix using the distance matrix and not the condensed vector matLink = hcluster.linkage(distMatrix) I get a different linkage matrix (the distances between the nodes are a lot larger and topology is slightly different)
Now I'm not sure whether this is because hcluster only works with condensed vectors or whether I'm making mistakes on the way there.
Thanks for your help!
I knocked up a quick random example similar to yours and experienced the same problem.
In the docstring it does say :
Performs hierarchical/agglomerative clustering on the
condensed distance matrix y. y must be a :math:{n \choose 2} sized
vector where n is the number of original observations paired
in the distance matrix.
However, having had a quick look at the code, it seems like the intent is for it to both work with vector shaped and matrix shaped code:
In hierachy.py there is a switch based upon the shape of the matrix.
It seems however that the key bit of info is in the function linkage's docstring:
- Q : ndarray
A condensed or redundant distance matrix. A condensed
distance matrix is a flat array containing the upper
triangular of the distance matrix. This is the form that
``pdist`` returns. Alternatively, a collection of
:math:`m` observation vectors in n dimensions may be passed as
a :math:`m` by :math:`n` array.
So I think that the interface doesn't allow the passing of a distance matrix.
Instead it thinks you are passing it m observation vectors in n dimensions .
Hence the difference in result?
Does that seem reasonable?
Else just take a look at the code itself I'm sure you'll be able to debug it and figure out why your examples are different.
Cheers
Matt