This data is from a real medical prediction task.
It comes from a study by Stamey et al in 1989
which tried to predict the amount of prostate specific antigens
based on a number of other clinical measures in 97 men who were about
to receive a radical prostatectomy. The output is the log prostate
specific antigen level (lpsa) normalized to lie between 0 and 1, 
and the inputs are (in order):

x1: log cancer volume   (lcavol)
x2: log prostate weight (lweight)
x3: age (age)
x4: log benign prostatic hyperplasia amount (lbph)
x5: seminal vesicle invasion (svi)
x6: log capsular penetration (lcp)
x7: average Gleason score (gleason)
x8: percent of Gleason scores in a certain range (pgg45)

There are 80 training cases and 17 test cases, 
with the split chosen at random.

For the ascii files, the LAST row gives the outputs,
and the first 8 rows give the inputs.