GSoC Weekly: LFortran and Computational Chemistry

Projects GSoC-2021
1

Hi All!
Apologies for the delayed start. I shall be linking to updates on my project here every Friday, and look forward to comments, suggestions, observations and other forms of engagement.

6 Likes
2

Thank you @rgoswami! Good job and looking forward.

1 Like
3

Updated with the second week’s progress report.

1 Like
4

Bump corresponding to the Week 3 update.

1 Like
5

Bump for Week 4.

1 Like
6

Early (short) update for Week 5.

1 Like
7

Short W6 update.

8

Strange, I can’t seem to edit the older post.

9

Bump for week 7.

1 Like
10

Bump for week 8

11

Week nine bump.

12

^ Not the final report, but the last weekly update in this format.

I managed to have some extra time this week and it was a lot of fun to invest it here :grin:

2 Likes
13

The last link leads to a 404 page.

14

Sorry about that, I just fixed it. CI issues… :sweat_smile:

15

How exactly did you generate this code in the post:

elemental real(dp) function kernel_dsin(x) result(res)
  real(dp), intent(in) :: x
  res = 2.087908370826698101212920076154545995e-37_dp + x * (0.999999999999999998041119383719347623_dp + x * (-1.064032149680587544373098154082500645e-35_dp + x * (-0.1666666666666666190025821826687444107_dp + x * (7.30686333395136116304293345061277226e-35_dp + x * (8.33333333333299311754500597882861296e-3_dp + x * (-1.943708819070258466622845490191457106e-35_dp + x * (-1.984126984115944978429391691458880356e-4_dp + x * (-5.30338548309302786712332804817767749e-34_dp + x * (2.75573192045805570474164007876304645e-6_dp + x * (1.185501640257404955632191608977614537e-33_dp + x * (-2.505210638108326593133314706727558551e-8_dp + x * (-1.130558221840288033840556502976715689e-33_dp + x * (1.605891864396986879258207676930409245e-10_dp + x * (5.55078059098189832214306988928325169e-34_dp + x * (-7.64251167703190749535921829507420458e-13_dp + x * (-1.378921764094824341145286520847883578e-34_dp + x * (2.716665406378575502010982726791586904e-15_dp + x * 1.374874330886318513437613275675104639e-35)))))))))))))))))
end function

The sollya code above that has different numbers. Also, we should check the accuracy of this approximation. In particular, the second coefficient will be just 1 in double precision.

1 Like
16

Sorry, I think I filled those in at different times; I’ve updated the post. Sometime soon I should be updating the representations with the even / odd coefficients correctly implemented.

1 Like
17

I forgot to cross post the final report, so…
Here it is!

LFortran and Compuational Chemistry

I had a fantastic time working on this and hope to continue my involvement :slight_smile:

6 Likes