I'm trying to implement double stochastic normalisation of an N x N x P tensor as described in Section 3.2 in Gong, CVPR 2019. This can be done easily in the N x N case using matrix operations but I am stuck with the 3D tensor case. What I have so far is
def doubly_stochastic_normalise(E):
"""E: n x n x f"""
E = E / torch.sum(E, dim=1, keepdim=True) # normalised across rows
F = E / torch.sum(E, dim=0, keepdim=True) # normalised across cols
E = torch.einsum('ijp,kjp->ikp', E, F)
return E
but I'm wondering if there is a method without einsum.
In this setting, you can always fall back to using torch.matmul (batched matrix multiplication to be more precise). However, this requires you to transpose the axis. Recall the matrix multiplication for two 3D inputs, in einsum notation, it gives us:
bik,bkj->bij
Notice how the k dimension gets reduces. To get to this setting, we need to transpose the inputs of the operator. In your case we have:
ijp ? kjp -> ikp
↓ ↓ ↑
pij # pjk -> pik
This translates to:
>>> (E.permute(2,0,1) # F.permute(2,1,0)).permute(1,2,0)
# ijp ➝ pij kjp ➝ pjk pik ➝ ikp
You can argue your method is not only shorter but also a lot more readable. I would therefore stick with torch.einsum. The reason why the einsum operator is so useful here is because you can perform axes transpositions on the fly.
Related
I am trying to approximate a line of best fit between multiple datasets, and display everything on one plot. This question addresses a similar notion, but the contents are in MatLab and, hence, not the same.
I have data from 4 different experiments that's composed of 146 values, the Y values represent changes in distance over time, the X value, which is represented by integer timesteps (1,2,3,...). The shape of my Y data is (4,146), as I've decided to keep all of it in a nested list, and the shape of my X data is (146,). I have the following set-up for my subplots:
x = [i for i in range(len(temp[0]))]
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.scatter(x,Y[0],c="blue", marker='.',linewidth=1)
ax1.scatter(x,Y[1],c="orange", marker='.',linewidth=1)
ax1.scatter(x,Y[2],c="green", marker='.',linewidth=1)
ax1.scatter(x,Y[3],c="purple", marker='.',linewidth=1)
z = np.polyfit(x,Y,3) # Throws an error because x,Y are not the same length
p = np.poly1d(z)
plt.plot(x, p(x))
I do not know how to fit a line of best fit between the scatter plots. numpy.polyfit documentation suggests that "Several data sets of sample points sharing the same x-coordinates can be fitted at once", but I have been unsuccessful thus far, and can only fit the line to one dataset. Is there a way that I can fit the line to all of the data sets? Should I use a different library entirely, like Seaborn?
Try to cast x and Y to a numpy arrays (I assume it is in a list). You can do this by using x = np.asarray(x). Now to fit on the data collectively, you can flatten the Y array using Y.flatten(). It transforms the shape from (n,N) to (n*N). And you can tile the x array n times to make a fit, this just copies the array n times into a new array so this will also become shape (n*N,). In this way you match the values form Y to corresponding values of x.
N = 10 # no. datapoints
n = 4 # no. experiments
# creating some dummy data
x = np.linspace(0,1, N) # shape (N,)
Y = np.random.normal(0,1,(n, N))
np.polyfit(np.tile(x, n), Y.flatten(), deg=3)
The polyfit function expects the Y array to be, in your case, (146, 4) rather than (4, 146), so you should pass it the transpose of Y, e.g.,
z = np.polyfit(x, Y.T, 3)
The poly1d function can only do one polynomial at a time, so you have to loop over the results from polyfit, e.g.,:
for res in z:
p = np.poly1d(res)
plt.plot(x, p(x))
I am reading the code for the spatial-temporal graph convolution operation here:
https://github.com/yysijie/st-gcn/blob/master/net/utils/tgcn.py and I'm having some difficulty understanding what is happening with an einsum operation. In particular
For x a tensor of shape (N, kernel_size, kc//kernel_size, t, v), where
kernel_size is typically 3, and lets say kc=64*kernel_size, t is the number of frames, say 64, and v the number of vertices, say 25. N is the batch size.
Now for a tensor A of shape (3, 25, 25) where each dimension is a filtering op on the graph vertices, an einsum is computed as:
x = torch.einsum('nkctv,kvw->nctw', (x, A))
I'm not sure how to interpret this expression. What I think it's saying is that for each batch element, for each channel c_i out of 64, sum each of the three matrices obtained by matrix multiplication of the (64, 25) feature map at that channel with the value of A[i]. Do I have this correct? The expression is a bit of a mouthful, and notation wise there seems to be a bit of a strange usage of kc as one variable name, but then decomposition of k as kernel size and c as the number of channels (192//3 = 64) in the expression for the einsum. Any insights appreciated.
Helps when you look closely at the notation:
nkctv for left side
kvw on the right side
nctw being the result
Missing from the result are:
k
v
These elements are summed together into a single value and squeezed, leaving us the resulting shape.
Something along the lines of (expanded shapes (added 1s) are broadcasted and sum per element):
left: (n, k, c, t, v, 1)
right: (1, k, 1, 1, v, w)
Now it goes (l, r for left and right):
torch.mul(l, r)
torch.sum(l, r, dim=(1, 4))
squeeze any singular dimensions
It is pretty hard to get, hence Einstein's summation helps in terms of thinking about resulting shapes “mixed” with each other, at least for me.
Y = torch.einsum('nkctv,kvw->nctw', (x, A)) means:
einsum interpretation on graph
For better understanding, I have replaced the x in left hand side with Y
I have three tensors v_1, v_2 and v_3, each of shape n x 3. And three tensors v_1', v_2' and v_3', also each of shape n x 3. I want to compute a tensor which stores n 3 x 3 matrices M_i, each solving the equation system
M_i * v_1_i = v_1_i'
M_i * v_2_i = v_2_i'
M_i * v_3_i = v_3_i'
It is guaranteed that this has one solution by construction. I just need the calculation for the rotation matrices. I tried torch.linalg.solve, but I can't figure out how to reshape the tensors correctly.
Thanks for your help.
I would like to implement the field-aware factorization model (FFM) in a vectorized way. In FFM, a prediction is made by the following equation
where w are the embeddings that depend on the feature and the field of the other feature. For more info, see equation (4) in FFM.
To do so, I have defined the following parameter:
import torch
W = torch.nn.Parameter(torch.Tensor(n_features, n_fields, n_factors), requires_grad=True)
Now, given an input x of size (batch_size, n_features), I want to be able to compute the previous equation. Here is my current (non-vectorized) implementation:
total_inter = torch.zeros(x.shape[0])
for i in range(n_features):
for j in range(i + 1, n_features):
temp1 = torch.mm(
x[:, i].unsqueeze(1),
W[i, feature2field[j], :].unsqueeze(0))
temp2 = torch.mm(
x[:, j].unsqueeze(1),
W[j, feature2field[i], :].unsqueeze(0))
total_inter += torch.sum(temp1 * temp2, dim=1)
Unsurprisingly, this implementation is horribly slow since n_features can easily be as large as 1000! Note however that most of the entries of x are 0. All inputs are appreciated!
Edit:
If it can help in any ways, here are some implementations of this model in PyTorch:
pytorch-fm
ctr_model_zoo
Unfortunately, I cannot figure out exactly how they have done it.
Additional update:
I can now obtain the product of x and W in a more efficient way by doing:
temp = torch.einsum('ij, jkl -> ijkl', x, W)
Thus, my loop is now:
total_inter = torch.zeros(x.shape[0])
for i in range(n_features):
for j in range(i + 1, n_features):
temp1 = temp[:, i, feature2field[j], :]
temp2 = temp[:, j, feature2field[i], :]
total_inter += 0.5 * torch.sum(temp1 * temp2, dim=1)
It is however still too long since this loop goes over for about 500 000 iterations.
Something that could potentially help you speed up the multiplication is using pytorch sparse tensors.
Also something that might work would be the following:
Create n arrays, one for each feature i that would hold its corresponding field factors in each row. e.g. for feature i = 0
[ W[0, feature2field[0], :],
W[0, feature2field[1], :],
W[0, feature2field[n], :]]
Then calculate the multiplication of those arrays, lets call them F, with X
R[i] = F[i] * X
So each element in R would hold the result of the multiplication, an array, of the F[i] with X.
Next you would multiply each R[i] with its transpose
R[i] = R[i] * R[i].T
Now you can do the summation in a loop like before
for i in range(n_features):
total_inter += torch.sum(R[i], dim=1)
Please take this with a grain of salt as i haven't tested it. In any case i think that it will point you in the right direction.
One problem that might occur is in the transpose multiplication in which each element will also be multiplied with itself and then be added in the sum. I don't think it will affect the classifier but in any case you can make the elements in the diagonal of the transpose and above 0 (including the diagonal).
Also although minor nevertheless please move the 1st unsqueeze operation outside of the nested for loop.
I hope it helps.
I am looking for a way to avoid the nested loops in the following snippet, where A and B are two-dimensional arrays, each of shape (m, n) with m, n beeing arbitray positive integers:
import numpy as np
m, n = 5, 2
a = randint(0, 10, (m, n))
b = randint(0, 10, (m, n))
out = np.empty((n, n))
for i in range(n):
for j in range(n):
out[i, j] = np.sum(A[:, i] + B[:, j])
The above logic is roughly equivalent to
np.einsum('ij,ik', A, B)
with the exception that einsum computes the sum of products.
Is there a way, equivalent to einsum, that computes a sum of sums? Or do I have to write an extension for this operation?
einsum needs to perform elementwise multiplication and then it does summing (optional). As such it might not be applicable/needed to solve our case. Read on!
Approach #1
We can leverage broadcasting such that the first axes are aligned
and second axis are elementwise summed after extending dimensions to 3D. Finally, we need summing along the first axis -
(A[:,:,None] + B[:,None,:]).sum(0)
Approach #2
We can simply do outer addition of columnar summations of each -
A.sum(0)[:,None] + B.sum(0)
Approach #3
And hence, bring in einsum -
np.einsum('ij->j',A)[:,None] + np.einsum('ij->j',B)
You can also use numpy.ufunc.outer, specifically here numpy.add.outer after summing along axis 0 as #Divakar mentioned in #approach 2
In [126]: numpy.add.outer(a.sum(0), b.sum(0))
Out[126]:
array([[54, 67],
[43, 56]])