https://docs.scipy.org/doc/numpy-1.15.0/reference/generated/numpy.dot.html

Dot product of two arrays. Specifically,

  • If both a and b are 1-D arrays, it is inner product of vectors (without complex conjugation).

  • If both a and b are 2-D arrays, it is matrix multiplication, but using matmul or a @ b is preferred.

  • If either a or b is 0-D (scalar), it is equivalent to multiply and using numpy.multiply(a, b) or a * b is preferred.

  • If a is an N-D array and b is a 1-D array, it is a sum product over the last axis of a and b.

  • If a is an N-D array and b is an M-D array (where M>=2), it is a sum product over the last axis of a and the second-to-last axis of b:

    dot(a, b)[i,j,k,m] = sum(a[i,j,:] * b[k,:,m])
What does the last two points mean?
My understanding is that 2-D arrays are a matrix. N-D array is a tensor. The aforementioned example is very clear. Both a and b are 3-D arrays.