Related
I want to create a new tensor z from two tensors, say x and y with dimensions [N_samples, S, N_feats] and [N_samples, T, N_feats] respectively. The aim is to combine both tensors on the 2nd dim by mixing the elements of the 2nd dim in a specific ordering, which is stored in a variable order with dim [N_samples, U].
The ordering is different for every sample and is basically which index to extract from which tensor. It looks like this for a given sample order[0] - [x_0, x_1, y_0, x_2, y_1, ... ], where the letter indicates the tensor and the number indicates the index of the 2nd dim. So z[0] would be
z[0] = [x[0, 0, :], x[0, 1, :], y[0, 0, :], x[0, 2, :], y[0, 1, :] ... ]
How would I achieve this? I've written something that uses torch.gather that tries to do this.
x = torch.rand((2, 4, 5))
y = torch.rand((2, 3, 5))
# new ordering of second dim
# positive means take (n-1)th element from x
# negative means take (n-1)th element from y
order = [[1, 2, -1, 3, -2, 4, 3],
[1, -1, -2, 2, 3, 4, -3]]
# simple concat for gather
combined = torch.cat([x, y], dim=1)
# add a zero padding on top of combined tensor to ease gather
zero = torch.zeros_like(x)[:, 1:2]
combined = torch.cat([zero, combined], dim=1)
def _create_index_for_gather(index, offset, n_feats):
new_index = [abs(i) + offset if i < 0 else i for i in index]
# need to repeat index for each dim for torch.gather
new_index = [[x] * n_feats for x in new_index]
return new_index
_, offset, n_feats = x.shape
index_for_gather = [_create_index_for_gather(i, offset, n_feats) for i in order]
z = combined.gather(dim=1, index=torch.tensor(index_for_gather))
Is there a more efficient way of doing this?
First things first: I'm relatively new to TensorFlow.
I'm trying to implement a custom layer in tensorflow.keras and I'm having relatively hard time when I try to achieve the following:
I've got 3 Tensors (x,y,z) of shape (?,49,3,3,32) [where ? is the batch size]
On each Tensor I compute the sum over the 3rd and 4th axes [thus I end up with 3 Tensors of shape (?,49,32)]
By doing an argmax (A)on the above 3 Tensors (?,49,32) I get a single (?,49,32) Tensor
Now I want to use this tensor to select slices from the initial x,y,z Tensors in the following form:
Each element in the last dimension of A corresponds to the selected Tensor.
(aka: 0 = X, 1 = Y, 2 = Z)
The index of the last dimension of A corresponds to the slice that I would like to extract from the Tensor last dimension.
I've tried to achieve the above using tf.gather but I had no luck. Then I tried using a series of tf.map_fn, which is ugly and computationally costly.
To simplify the above:
let's say we've got an A array of shape (3,3,3,32). Then the numpy equivalent of what I try to achieve is this:
import numpy as np
x = np.random.rand(3,3,32)
y = np.random.rand(3,3,32)
z = np.random.rand(3,3,32)
x_sums = np.sum(np.sum(x,axis=0),0);
y_sums = np.sum(np.sum(y,axis=0),0);
z_sums = np.sum(np.sum(z,axis=0),0);
max_sums = np.argmax([x_sums,y_sums,z_sums],0)
A = np.array([x,y,z])
tmp = []
for i in range(0,len(max_sums)):
tmp.append(A[max_sums[i],:,:,i)
output = np.transpose(np.stack(tmp))
Any suggestions?
ps: I tried tf.gather_nd but I had no luck
This is how you can do something like that with tf.gather_nd:
import tensorflow as tf
# Make example data
tf.random.set_seed(0)
b = 10 # Batch size
x = tf.random.uniform((b, 49, 3, 3, 32))
y = tf.random.uniform((b, 49, 3, 3, 32))
z = tf.random.uniform((b, 49, 3, 3, 32))
# Stack tensors together
data = tf.stack([x, y, z], axis=2)
# Put reduction axes last
data_t = tf.transpose(data, (0, 1, 5, 2, 3, 4))
# Reduce
s = tf.reduce_sum(data_t, axis=(4, 5))
# Find largest sums
idx = tf.argmax(s, 3)
# Make gather indices
data_shape = tf.shape(data_t, idx.dtype)
bb, ii, jj = tf.meshgrid(*(tf.range(data_shape[i]) for i in range(3)), indexing='ij')
# Gather result
output_t = tf.gather_nd(data_t, tf.stack([bb, ii, jj, idx], axis=-1))
# Reorder axes
output = tf.transpose(output_t, (0, 1, 3, 4, 2))
print(output.shape)
# TensorShape([10, 49, 3, 3, 32])
The Problem:
I want to calculate the dot product of a very large set of data. I am able to do this in a nested for-loop, but this is way too slow.
Here is a small example:
import numpy as np
points = np.array([[0.5, 2, 3, 5.5, 8, 11], [1, 2, -1.5, 0.5, 4, 5]])
lines = np.array([[0, 2, 4, 6, 10, 10, 0, 0], [0, 0, 0, 0, 0, 4, 4, 0]])
x1 = lines[0][0:-1]
y1 = lines[1][0:-1]
L1 = np.asarray([x1, y1])
# calculate the relative length of the projection
# of each point onto each line
a = np.diff(lines)
b = points[:,:,None] - L1[:,None,:]
print(a.shape)
print(b.shape)
[rows, cols, pages] = np.shape(b)
Z = np.zeros((cols, pages))
for k in range(cols):
for l in range(pages):
Z[k][l] = a[0][l]*b[0][k][l] + a[1][l]*b[1][k][l]
N = np.linalg.norm(a, axis=0)**2
relativeProjectionLength = np.squeeze(np.asarray(Z/N))
In this example, the first two dimensions of both a and b represent the x- and y-coordinates that I need for the dot product.
The shape of a is (2,7) and b has (2,6,7). Since the dot product reduces the first dimension I would expect the result to be of the shape (6,7). How can I calculate this without the slow loops?
What I have tried:
I think that numpy.dot with correct broadcasting could do the job, however I have trouble setting up the dimensions correctly.
a = a[:, None, :]
Z = np.dot(a,b)
This on gives me the following error:
shapes (2,1,7) and (2,6,7) not aligned: 7 (dim 2) != 6 (dim 1)
You can use np.einsum -
np.einsum('ij,ikj->kj',a,b)
Explanation :
Keep the last axes aligned for the two inputs.
Sum-reduce the first from those.
Let the rest stay, which is the second axis of b.
Usual rules on whether to use einsum or stick to a loopy-dot based method apply here.
numpy.dot does not reduce the first dimension. From the docs:
For N dimensions it is a sum product over the last axis of a and the second-to-last of b:
dot(a, b)[i,j,k,m] = sum(a[i,j,:] * b[k,:,m])
That is exactly what the error is telling you: it is attempting to match axis 2 in the first vector to axis 1 in the second.
You can fix this using numpy.rollaxis or better yet numpy.moveaxis. Instead of a = a[:, None, :], do
a = np.movesxis(a, 0, -1)
b = np.moveaxis(b, 0, -2)
Z = np.dot(a, b)
Better yet, you can construct your arrays to have the correct shape up front. For example, transpose lines and do a = np.diff(lines, axis=0).
I want to compute the pairwise square distance of a batch of feature in Tensorflow. I have a simple implementation using + and * operations by
tiling the original tensor :
def pairwise_l2_norm2(x, y, scope=None):
with tf.op_scope([x, y], scope, 'pairwise_l2_norm2'):
size_x = tf.shape(x)[0]
size_y = tf.shape(y)[0]
xx = tf.expand_dims(x, -1)
xx = tf.tile(xx, tf.pack([1, 1, size_y]))
yy = tf.expand_dims(y, -1)
yy = tf.tile(yy, tf.pack([1, 1, size_x]))
yy = tf.transpose(yy, perm=[2, 1, 0])
diff = tf.sub(xx, yy)
square_diff = tf.square(diff)
square_dist = tf.reduce_sum(square_diff, 1)
return square_dist
This function takes as input two matrices of size (m,d) and (n,d) and compute the squared distance between each row vector. The output is a matrix of size (m,n) with element 'd_ij = dist(x_i, y_j)'.
The problem is that I have a large batch and high dim features 'm, n, d' replicating the tensor consume a lot of memory.
I'm looking for another way to implement this without increasing the memory usage and just only store the final distance tensor. Kind of double looping the original tensor.
You can use some linear algebra to turn it into matrix ops. Note that what you need matrix D where a[i] is the ith row of your original matrix and
D[i,j] = (a[i]-a[j])(a[i]-a[j])'
You can rewrite that into
D[i,j] = r[i] - 2 a[i]a[j]' + r[j]
Where r[i] is squared norm of ith row of the original matrix.
In a system that supports standard broadcasting rules you can treat r as a column vector and write D as
D = r - 2 A A' + r'
In TensorFlow you could write this as
A = tf.constant([[1, 1], [2, 2], [3, 3]])
r = tf.reduce_sum(A*A, 1)
# turn r into column vector
r = tf.reshape(r, [-1, 1])
D = r - 2*tf.matmul(A, tf.transpose(A)) + tf.transpose(r)
sess = tf.Session()
sess.run(D)
result
array([[0, 2, 8],
[2, 0, 2],
[8, 2, 0]], dtype=int32)
Using squared_difference:
def squared_dist(A):
expanded_a = tf.expand_dims(A, 1)
expanded_b = tf.expand_dims(A, 0)
distances = tf.reduce_sum(tf.squared_difference(expanded_a, expanded_b), 2)
return distances
One thing I noticed is that this solution using tf.squared_difference gives me out of memory (OOM) for very large vectors, while the approach by #YaroslavBulatov doesn't. So, I think decomposing the operation yields a smaller memory footprint (which I thought squared_difference would handle better under the hood).
Here is a more general solution for two tensors of coordinates A and B:
def squared_dist(A, B):
assert A.shape.as_list() == B.shape.as_list()
row_norms_A = tf.reduce_sum(tf.square(A), axis=1)
row_norms_A = tf.reshape(row_norms_A, [-1, 1]) # Column vector.
row_norms_B = tf.reduce_sum(tf.square(B), axis=1)
row_norms_B = tf.reshape(row_norms_B, [1, -1]) # Row vector.
return row_norms_A - 2 * tf.matmul(A, tf.transpose(B)) + row_norms_B
Note that this is the square distance. If you want to change this to the Euclidean distance, perform a tf.sqrt on the result. If you want to do that, don't forget to add a small constant to compensate for the floating point instabilities: dist = tf.sqrt(squared_dist(A, B) + 1e-6).
If you want compute other method , then change the order of the tf modules.
def compute_euclidean_distance(x, y):
size_x = x.shape.dims[0]
size_y = y.shape.dims[0]
for i in range(size_x):
tile_one = tf.reshape(tf.tile(x[i], [size_y]), [size_y, -1])
eu_one = tf.expand_dims(tf.sqrt(tf.reduce_sum(tf.pow(tf.subtract(tile_one, y), 2), axis=1)), axis=0)
if i == 0:
d = eu_one
else:
d = tf.concat([d, eu_one], axis=0)
return d
The general solution to this question is being worked on in this github issue, but I was wondering if there are workarounds using tf.gather (or something else) to achieve array indexing using a multi-index. One solution I came up with was to broadcast multiply each index in the multi-idx with the cumulative product of the tensor shape, which produces indices suitable for indexing the flattened tensor:
import tensorflow as tf
import numpy as np
def __cumprod(l):
# Get the length and make a copy
ll = len(l)
l = [v for v in l]
# Reverse cumulative product
for i in range(ll-1):
l[ll-i-2] *= l[ll-i-1]
return l
def ravel_multi_index(tensor, multi_idx):
"""
Returns a tensor suitable for use as the index
on a gather operation on argument tensor.
"""
if not isinstance(tensor, (tf.Variable, tf.Tensor)):
raise TypeError('tensor should be a tf.Variable')
if not isinstance(multi_idx, list):
multi_idx = [multi_idx]
# Shape of the tensor in ints
shape = [i.value for i in tensor.get_shape()]
if len(shape) != len(multi_idx):
raise ValueError("Tensor rank is different "
"from the multi_idx length.")
# Work out the shape of each tensor in the multi_idx
idx_shape = [tuple(j.value for j in i.get_shape()) for i in multi_idx]
# Ensure that each multi_idx tensor is length 1
assert all(len(i) == 1 for i in idx_shape)
# Create a list of reshaped indices. New shape will be
# [1, 1, dim[0], 1] for the 3rd index in multi_idx
# for example.
reshaped_idx = [tf.reshape(idx, [1 if i !=j else dim[0]
for j in range(len(shape))])
for i, (idx, dim)
in enumerate(zip(multi_idx, idx_shape))]
# Figure out the base indices for each dimension
base = __cumprod(shape)
# Now multiply base indices by each reshaped index
# to produce the flat index
return (sum(b*s for b, s in zip(base[1:], reshaped_idx[:-1]))
+ reshaped_idx[-1])
# Shape and slice starts and sizes
shape = (Z, Y, X) = 4, 5, 6
Z0, Y0, X0 = 1, 1, 1
ZS, YS, XS = 3, 3, 4
# Numpy matrix and index
M = np.random.random(size=shape)
idx = [
np.arange(Z0, Z0+ZS).reshape(ZS,1,1),
np.arange(Y0, Y0+YS).reshape(1,YS,1),
np.arange(X0, X0+XS).reshape(1,1,XS),
]
# Tensorflow matrix and indices
TM = tf.Variable(M)
TF_flat_idx = ravel_multi_index(TM, [
tf.range(Z0, Z0+ZS),
tf.range(Y0, Y0+YS),
tf.range(X0, X0+XS)])
TF_data = tf.gather(tf.reshape(TM,[-1]), TF_flat_idx)
with tf.Session() as S:
S.run(tf.initialize_all_variables())
# Obtain data via flat indexing
data = S.run(TF_data)
# Check that it agrees with data obtained
# by numpy smart indexing
assert np.all(data == M[idx])
However, this only works on tensors of rank 3 due to this (current) limitation limiting broadcasts to tensors of rank 3.
At the moment I can only think of doing a chained gather, transpose, gather, transpose, gather, but this is unlikely to be efficient. e.g.
shape = (8, 9, 10)
A = tf.random_normal(shape)
data = tf.gather(tf.transpose(tf.gather(A, [1, 3]), [1,0,2]), ...)
Any ideas?
It sounds like you want gather_nd.