hamming | Hamming distance and bit counting primitives in Go | Learning library

I have finally done this, which seems to work well:

A vectorized (read "much faster") solution:

I think you've misunderstood the M_2_PI constant in your GenerateSinWave function. M_2_PI is defined as 2.0 / PI. You should be using 2 * M_PI instead.

This mistake will mean that your generated signal has a frequency of only around 45 Hz. This should be close to the output frequencies you are seeing.

The same constant needs correcting in your HammingWindow function too.

Every m-sized subset of the n dimensions is an "orientation" with those m dimensions "free". For each orientation, you can generate a face by generating all 2^m combinations of the m coordinates in the m free dimensions, while holding the coordinates for the other dimensions fixed. Each of the 2^n-m combinations of coordinates for the other dimensions is the "position" of a different face.

The number of orientations is C(n,m) = n!/m!(n-m)!, so you should end up with C(n,m) * 2^n-m faces overall.

The number of edges in a cube, for example = C(3,1) * 2² = 3 * 4 = 12.

The number of faces on a cube is C(3,2) * 2 = 3 * 2 = 6.

It's not too difficult to generate all faces:

• For each orientation, determine the free and fixed dimensions
  • Count in binary using the fixed dimensions to find all the face positions
    • For each face, count in binary over the free dimensions to generate its vertices.

If dim(M1) and dim(M2) are identical, then you can simply do:

I think if you reshape your image to be 4D, you should be good to go. So, after image = np.expand_dims(image, axis=0), just run image = image.reshape(image.shape + (1,)):

Make sure to save your images as .png files. .jpg files are lossy, and don't compress hard lines very well, since they are designed to store regular photos. If you look at the raw photo data, you will see a sharp transition between zones when you interpolated with NEAREST, but since the image is saved as .jpg, that all goes out the window.

Here's what the raw data is when you downsize with NEAREST:

Here's what the .jpg data looks like:

You can see the transition if filled with noise.

The function pairwise_distances can take in a matrix, so it might be easier to just provide the features in a year as a matrix, get back a pairwise matrix of distances and just subset on the comparisons we need. For example, a dataset like yours:

Some small points: