Commit graph

551 commits

Author SHA1 Message Date
Fredrik Johansson
0244fe0e46 Fix evil bug in floating-point arithmetic and strengthen test code.
When rounding up, the error bound sometimes came out as 1/2 ulp instead of 1 ulp due to the mantissa rounding up to the next power of two, without a corresponding exponent correction. Fortunately, this should not affect ball arithmetic (which always rounds down).
2014-04-18 16:19:53 +02:00
Fredrik Johansson
ef5998f085 merge 2014-04-18 16:19:32 +02:00
Fredrik Johansson
5697316463 simplify interface 2014-04-18 16:17:42 +02:00
Fredrik Johansson
24c2f11721 some code reorganisation; optimise fmpr_mul_si/ui/fmpz 2014-04-18 16:17:22 +02:00
Fredrik Johansson
462cd887ff use doubles for faster error bounds in fmprb_poly_mullow_block2 2014-04-16 22:57:15 +02:00
Fredrik Johansson
8c71a3c8dc more accurate version of fmprb_poly_mullow_block (two full poly muls for error bounds) 2014-04-15 17:07:02 +02:00
Fredrik Johansson
bd4696419b Merge remote branch 'origin/master' 2014-04-15 17:02:14 +02:00
Fredrik Johansson
e310ff122a remove an unused variable and rename an internal function to avoid namespace conflict with flint 2014-04-15 17:01:45 +02:00
Fredrik Johansson
b0a65420ac minor doc tweaks 2014-04-05 12:01:45 +02:00
Fredrik Johansson
009a0eb918 respect use_doubles 2014-03-29 18:39:15 +01:00
Fredrik Johansson
f31b7c29f6 use the new fmpz_poly_cos_minpoly in flint instead of elefun_cos_minpoly 2014-03-29 07:42:16 +01:00
Fredrik Johansson
3e1f679259 include multithreaded partition function and update documentation 2014-03-28 17:00:10 +01:00
Fredrik Johansson
909e929f07 fix a comment 2014-03-28 15:45:06 +01:00
Fredrik Johansson
1959ebe699 typo 2014-03-04 12:55:38 +01:00
Fredrik Johansson
72059632d1 document new exponential function code 2014-03-04 12:39:12 +01:00
Fredrik Johansson
61e0c0c2ff support partition function of bignum n (work in progress) 2014-03-03 16:32:29 +01:00
Fredrik Johansson
2810e26428 fix test code bug on 32 bit 2014-02-24 12:56:22 +01:00
Fredrik Johansson
11d5176ddb new exp bit-burst implementation 2014-02-23 17:55:53 +01:00
Fredrik Johansson
f002340f2a Merge remote branch 'origin/master' 2014-02-23 16:05:04 +01:00
Fredrik Johansson
b45b16584c new binary splitting code for exponentials 2014-02-23 16:04:36 +01:00
Fredrik Johansson
d7e7ca65ac make exp Taylor bound work at higher precision (tentative implementation, using math.h) 2014-02-23 15:26:22 +01:00
Fredrik Johansson
4f71268798 add a todo item 2014-01-23 16:24:03 +01:00
Fredrik Johansson
afa3a98d7f remove the holonomic module (too experimental for now; not used anywhere) 2014-01-15 12:54:28 +01:00
Fredrik Johansson
a9c2951dc7 change a ptr -> srcptr 2014-01-15 12:08:49 +01:00
Fredrik Johansson
27ebe64238 don't require zero-padded input in series reversion 2014-01-15 11:40:11 +01:00
Fredrik Johansson
e6ecc103cf improve rising factorial series for length 1, 2; fix a bug and add missing test file 2013-12-29 22:04:00 +01:00
Fredrik Johansson
477bcebf1e small code cleanup 2013-12-29 19:29:42 +01:00
Fredrik Johansson
c3d61204e8 faster digamma function argument reduction 2013-12-29 18:07:56 +01:00
Fredrik Johansson
7a8f9038e9 mark this 1.0.0 2013-12-21 22:39:35 +01:00
Fredrik Johansson
feb195cfb4 add poly_roots example program 2013-12-16 13:04:09 +01:00
Fredrik Johansson
366124b35a some documentation edits 2013-12-03 16:30:33 +01:00
Fredrik Johansson
b537658904 simplify const_euler and use the published bound 2013-12-03 16:06:25 +01:00
Fredrik Johansson
6a8d3772da slight improvement to exp_series_basecase for large n 2013-11-24 02:40:57 +01:00
Fredrik Johansson
8d80118bbd add fmprb_inv and replace uses of fmprb_ui_div 2013-11-12 15:25:50 +01:00
Fredrik Johansson
8ff94cff7d improve wording 2013-11-11 19:24:42 +01:00
Fredrik Johansson
b78e176b0d implement matrix exponentials 2013-11-11 18:50:53 +01:00
Fredrik Johansson
7ce5df8690 add matrix infinity norm functions 2013-11-11 18:21:55 +01:00
Fredrik Johansson
5a3439de89 fix an inline declaration; some cleanup 2013-11-11 18:13:56 +01:00
Fredrik Johansson
2b60af9281 add missing matrix-scalar functions 2013-11-11 17:51:37 +01:00
Fredrik Johansson
6bd2b7078a add missing matrix contains/overlaps methods 2013-11-11 17:36:45 +01:00
Fredrik Johansson
38231fc1f3 re-enable a case 2013-10-26 21:51:17 +02:00
Fredrik Johansson
28814f40c2 special case +/- 2 in fmpcb_root 2013-10-22 11:41:58 +02:00
Fredrik Johansson
7d663e14f7 remove unnecessary include 2013-10-21 17:59:32 +02:00
Fredrik Johansson
9ded02956b add fmpcb_rsqrt (complex reciprocal square root) 2013-10-21 17:58:42 +02:00
Fredrik Johansson
9affdac67e improve special cases in sqrt/rsqrt series 2013-10-21 17:08:17 +02:00
Fredrik Johansson
c2d8b6dac8 slightly improve sin/cos/tan_series 2013-10-21 16:45:11 +02:00
Fredrik Johansson
94120b0ddd improve log_series for special-form input 2013-10-21 15:30:19 +02:00
Fredrik Johansson
be6a8ee46f small cleanup 2013-10-21 14:58:09 +02:00
Fredrik Johansson
852529f511 improve exp_series for medium-size and special-form input 2013-10-21 14:52:49 +02:00
Fredrik Johansson
ba9d629ad0 add missing documentation entries for fmpcb_pow_si and fmpcb_pow_fmprb 2013-10-14 15:02:08 +02:00