Cant figure out what is wrong with my compression

[alykhanb96]

Wrote this code to compress:

def compress(n):
  newU = u[0:768, 0:n]
  newV = v[0:n, 0:1024]
  u_n= newU
  v_n= newV
  s_n= np.eye(n, dtype=float) * np.diag(s[0:n])

Each time i run it I run it this way:

compress(500)
pre = np.matmul(u_n, s_n)
face_compressed_500= np.matmul(pre, v_n)
plt.imshow(face_compressed_500, cmap=plt.cm.gray)

compress(300)
pre = np.matmul(u_n, s_n)
face_compressed_300= np.matmul(pre, v_n)
plt.imshow(face_compressed_300, cmap=plt.cm.gray)

But i get the exact same picture, furthermore my storage size is only 200 apart

[alykhanb96]

@jmbejara can you take a look at this if you have a moment. Thanks so much

[jmbejara]

It seems like the main point is that you're not returning anything from your compress function. The variables you define in the function disappear when the function finishes. To understand this, take some time to go through the discussion here: https://github.com/jmbejara/comp-econ-sp18/issues/2

Some other points:

• Use @ for matrix multiplications. It makes the code more readable.
• u[0:768, 0:n] can be written better like this: u[:,0:n]. Similar comment with v.
• s_n= np.eye(n, dtype=float) * np.diag(s[0:n]) seems redundant.

Let me know if this makes sense. Be sure to stop by the office hours at 1:45 as well. Hope this helps!