Understanding tensordot

by

The idea with tensordot is pretty simple – We input the arrays and the respective axes along which the sum-reductions are intended. The axes that take part in sum-reduction are removed in the output and all of the remaining axes from the input arrays are spread-out as different axes in the output keeping the order in which the input arrays are fed.

Let’s look at few sample cases with one and two axes of sum-reductions and also swap the input places and see how the order is kept in the output.

I. One axis of sum-reduction

Inputs :

 In [7]: A = np.random.randint(2, size=(2, 6, 5))
   ...:  B = np.random.randint(2, size=(3, 2, 4))
   ...:

Case #1:

In [9]: np.tensordot(A, B, axes=((0),(1))).shape
Out[9]: (6, 5, 3, 4)

A : (2, 6, 5) -> reduction of axis=0
B : (3, 2, 4) -> reduction of axis=1

Output : `(2, 6, 5)`, `(3, 2, 4)` ===(2 gone)==> `(6,5)` + `(3,4)` => `(6,5,3,4)`

Case #2 (same as case #1 but the inputs are fed swapped): 

In [8]: np.tensordot(B, A, axes=((1),(0))).shape
Out[8]: (3, 4, 6, 5)

B : (3, 2, 4) -> reduction of axis=1
A : (2, 6, 5) -> reduction of axis=0

Output : `(3, 2, 4)`, `(2, 6, 5)` ===(2 gone)==> `(3,4)` + `(6,5)` => `(3,4,6,5)`.

II. Two axes of sum-reduction

Inputs :

In [11]: A = np.random.randint(2, size=(2, 3, 5))
    ...: B = np.random.randint(2, size=(3, 2, 4))
    ...:

Case #1:

In [12]: np.tensordot(A, B, axes=((0,1),(1,0))).shape
Out[12]: (5, 4)

A : (2, 3, 5) -> reduction of axis=(0,1)
B : (3, 2, 4) -> reduction of axis=(1,0)

Output : `(2, 3, 5)`, `(3, 2, 4)` ===(2,3 gone)==> `(5)` + `(4)` => `(5,4)`

Case #2:

In [14]: np.tensordot(B, A, axes=((1,0),(0,1))).shape
Out[14]: (4, 5)

B : (3, 2, 4) -> reduction of axis=(1,0)
A : (2, 3, 5) -> reduction of axis=(0,1)

Output : `(3, 2, 4)`, `(2, 3, 5)` ===(2,3 gone)==> `(4)` + `(5)` => `(4,5)`

We can extend this to as many axes as possible.

More Related Contents:

  1. “Cloning” row or column vectors
  2. AttributeError: ‘Tensor’ object has no attribute ‘numpy’
  3. Link ATLAS/MKL to an installed Numpy
  4. Fitting a line in 3D
  5. numpy arbitrary precision linear algebra
  6. Why does numpy.linalg.solve() offer more precise matrix inversions than numpy.linalg.inv()?
  7. Equivalent of `polyfit` for a 2D polynomial in Python
  8. How to find linearly independent rows from a matrix
  9. Efficient dot products of large memory-mapped arrays
  10. Convert a 3D array to RGB valued array [closed]
  11. scipy: savefig without frames, axes, only content
  12. Interpolate NaN values in a numpy array
  13. Numpy ‘smart’ symmetric matrix
  14. Fitting a 2D Gaussian function using scipy.optimize.curve_fit – ValueError and minpack.error
  15. ‘DataFrame’ object has no attribute ‘sort’
  16. How to get a list of all indices of repeated elements in a numpy array
  17. N-D version of itertools.combinations in numpy
  18. figure of imshow() is too small
  19. python recursive vectorization with timeseries
  20. Difference between np.random.seed() and np.random.RandomState()
  21. Flattening a list of NumPy arrays?
  22. Fastest pairwise distance metric in python
  23. Is there an analysis speed or memory usage advantage to using HDF5 for large array storage (instead of flat binary files)?
  24. python tilde unary operator as negation numpy bool array
  25. How to shade region under the curve in matplotlib
  26. Randomly select from numpy array
  27. How to use NumPy array with ctypes?
  28. How to use numpy to get the cumulative count by unique values in linear time?
  29. Annotate Time Series plot
  30. Numpy `ValueError: operands could not be broadcast together with shape …`

Leave a Comment