ae104919e0
Still not sure whether it's better to calculate rref[pivot_row][col] * (rref[row][pivot_col] / pivot_val) (option 1) vs. rref[pivot_row][col] * rref[row][pivot_col] / pivot_val (option 2) (i.e. a * b / c vs. "a * (b/c)) in terms of floating point error. But I think option 1 (current commit) is better, since the scale factor (rref[row][pivot_col] / pivot_val) is always <= 1 here (I think).
10 lines
345 B
Plaintext
10 lines
345 B
Plaintext
- guf_stack, guf_queue, guf_ringbuf
|
|
|
|
- no guf_init.h
|
|
|
|
- linalg: float precision question += elem * -val / pivot_val vs elem * (-val / pivot_val)
|
|
|
|
- unicode normalisation
|
|
- track allocs for test (implement alloc tracker)
|
|
- handle right-to-left text properly
|
|
- fix 32-bit dict (and add 32/64 bit defs and 32/64-bit platform detection in common.h) |