arguments: m:Matrix threshold:RingElement flag:int >> A:SparseMutableMatrix LLLstate:SparseMutableMatrix
Setup the computation of the LLL reduction of the matrix 'm' (over ZZ). There is no engine LLL computation object. Instead, the answer will appear in 'A' (after using ggLLLcalc), using the internal state 'LLLstate'. The threshold should be an element of ZZ or QQ, in the range (1/4, 1]. If flag is non-zero, then the change of basis column multiplier matrix for 'A' is set. After the computation is complete (or before, to view the partial answer), this multiplier matrix B can be obtained using gggetColChange on A. The final answers satisfy: A = m*B. For more information about LLL reduction, see H. Cohen, A Course in Computational Algebraic Number Theory, Springer Lecture notes in Math, 138.