List of datasets

• For Q2 - Mean/median paradox: Financial amounts here.
  • About 1M (2**20) entries, plus a header-line.
• For Q3 - Similarities and SVD: the 25x25 matrix of similarity scores is here
  • 25 lines with 25 entries each - no header.
  • The similarity scores should have 6 digits of accuracy.
• For Q4: Fourier and denoising. The sound-like signal is here.
  • 1024 lines, with one value per line - no header.