Once a lambda is found (or a candidate chosen), compute probs and validate.

Workaround since sprintf is inconsistent in double rounding for different implementations.

two corresponding arrays of probabilities d_array_p [0,1][0...d_arrayCapacity - 1]

* Returns the current resident set size (physical memory use) measured * in bytes, or zero if the value cannot be determined on this OS. */

minimum expectation (exp (theta * score))

function for updating dynamic programming values

probability of "letters" p2_ [0...dimMatrix_), the second (j) set of letter-probabilities

step standard deviation for associated random walk

renewal span parameter

step standard deviation for random walk

score matrix [0...dimMatrix_)[0...dimMatrix_)

probability of "letters" d_p_p [0...dimMatrix_)

* Returns the peak (maximum so far) resident set size (physical * memory use) measured in bytes, or zero if the value cannot be * determined on this

scores in increasing order

theta for minimum expectation (exp (theta * score))

#(distinct values) of scores & probabilities (which are paired)

#(distinct values) of scores & probabilities (which are paired)

step mean for random walk

step mean for associated random walk (relative entropy)

c for random walk : exponential prefactor (global alignment)

span

# (different values)

lower limit for long int values in the array (an offset)