Timeline for Python 3 - Fibonacci Implementation
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 22, 2017 at 16:07 | comment | added | Eric Duminil | @Snowbody you're correct | |
| Dec 22, 2017 at 16:05 | comment | added | 13ros27 | Does anyone know how to get a more accurate value of Phi | |
| Dec 22, 2017 at 16:04 | comment | added | Snowbody | @EricDuminil that's not it. Once n is large enough the second term is near-zero and has no contribution to the formula. The bigger problem is that the small inaccuracy in Phi due to limited precision gets magnified when exponentiated. Phi is actually off the correct value by sys.float_info.epsilon or so. On my machine Phi is real Phi + around 4e-16 = real Phi*(1+2.45e-16). So n=72 seems about the right place. | |
| Dec 20, 2017 at 12:35 | history | edited | 13ros27 | CC BY-SA 3.0 | Corrected spelling |
| Dec 20, 2017 at 12:35 | comment | added | 13ros27 | It is the value of Phi, it loses accuracy the larger the number you are finding | |
| Dec 20, 2017 at 12:34 | comment | added | Eric Duminil | The main problem is that this calculates the difference of a huge number and a very small number. There's still a typo in your function name BTW. | |
| Dec 20, 2017 at 10:39 | history | edited | 13ros27 | CC BY-SA 3.0 | Corrected spelling |
| Dec 20, 2017 at 10:33 | comment | added | Eric Duminil | It's Fibonacci, not Fibonnacci or Fibonnaci. While it's good to know that this formula exists and is correct in theory, it's already returning wrong values for n=72. | |
| Dec 20, 2017 at 10:22 | history | edited | Graipher | CC BY-SA 3.0 | added 1 character in body; edited body |
| Dec 20, 2017 at 10:10 | history | edited | 13ros27 | CC BY-SA 3.0 | Added a note |
| Dec 20, 2017 at 10:09 | comment | added | 13ros27 | Good point, I will add it | |
| Dec 20, 2017 at 9:37 | history | edited | Konrad Rudolph | CC BY-SA 3.0 | typo, formatting (PEP8) |
| Dec 20, 2017 at 9:24 | comment | added | Ziyad Edher | Would it not be better to once-compute math.sqrt(5) at the start, if we are looking to get the most performance, or does it get automatically get cached? | |
| Dec 20, 2017 at 9:07 | history | edited | hjpotter92 | CC BY-SA 3.0 | use tex for mathematical formula |
| Dec 20, 2017 at 8:41 | history | edited | 13ros27 | CC BY-SA 3.0 | Corrected code |
| Dec 20, 2017 at 7:58 | history | answered | 13ros27 | CC BY-SA 3.0 |