brain decoding results

First stab at decoding P01 data onto InferSent embeddings (dim 4096).
This is promising performance, since I haven't touched either the imaging data or the encodings since they were produced. (The regression is ~20,000 dim -> 4096 dim – definitely can improve with some preprocessing.)
Compare with original results from the problem set later on in this gist.

INFO:__main__:Loaded encodings of size (384, 4096).
INFO:__main__:Loaded subject data/P01 data.
INFO:__main__:Trained classifier for subject data/P01.
Fold  0: min  4.0       mean  96.5      med  41.0       max  350.0
Fold  1: min  0.0       mean  131.9     med  68.0       max  323.0
Fold  2: min  2.0       mean  144.3     med  129.0      max  366.0
Fold  3: min  15.0      mean  166.5     med  165.0      max  365.0
Fold  4: min  0.0       mean  116.0     med  130.0      max  374.0
Fold  5: min  0.0       mean  76.6      med  39.0       max  305.0
Fold  6: min  1.0       mean  75.1      med  42.0       max  299.0
Fold  7: min  1.0       mean  72.0      med  29.0       max  344.0
Fold  8: min  0.0       mean  87.5      med  23.0       max  339.0
Fold  9: min  0.0       mean  83.1      med  54.0       max  296.0
Fold  10: min  10.0     mean  124.8     med  64.0       max  328.0
Fold  11: min  9.0      mean  135.5     med  108.0      max  343.0
Fold  12: min  0.0      mean  79.6      med  47.0       max  285.0
Fold  13: min  1.0      mean  69.4      med  46.0       max  260.0
Fold  14: min  9.0      mean  109.6     med  65.0       max  361.0
Fold  15: min  5.0      mean  114.4     med  104.0      max  327.0
Fold  16: min  1.0      mean  82.4      med  46.0       max  356.0
Fold  17: min  0.0      mean  80.2      med  65.0       max  370.0


-------------

Results from the original problem set, using their semantic vectors and imaging data:
(I think the imaging data here was preprocessed somehow, need to re-download it and analyze)

Fold  0: min  7.0	mean  65.7	med  54.5	max  179.0
Fold  1: min  7.0	mean  61.3	med  59.0	max  134.0
Fold  2: min  9.0	mean  59.4	med  39.0	max  127.0
Fold  3: min  8.0	mean  69.6	med  68.5	max  177.0
Fold  4: min  11.0	mean  80.3	med  78.5	max  170.0
Fold  5: min  0.0	mean  73.5	med  74.5	max  163.0
Fold  6: min  9.0	mean  76.0	med  55.5	max  167.0
Fold  7: min  0.0	mean  45.7	med  38.5	max  112.0
Fold  8: min  47.0	mean  104.1	med  119.0	max  145.0
Fold  9: min  3.0	mean  38.1	med  35.0	max  114.0
Fold  10: min  6.0	mean  64.6	med  57.0	max  158.0
Fold  11: min  8.0	mean  55.5	med  48.5	max  134.0
Fold  12: min  7.0	mean  35.9	med  25.0	max  116.0
Fold  13: min  13.0	mean  65.0	med  46.5	max  156.0
Fold  14: min  11.0	mean  40.7	med  31.0	max  151.0
Fold  15: min  4.0	mean  35.8	med  34.0	max  66.0
Fold  16: min  0.0	mean  38.7	med  32.5	max  97.0
Fold  17: min  12.0	mean  86.5	med  106.5	max  157.0