Out-source n!/k! to Combinatorics.jl #7
Conversation
jagot
commented
Jul 8, 2020
jagot
commented
| @@ -50,7 +50,7 @@ function radial_dipole_moment(a::Orbital, b::Orbital) | |||
| n == n′ && return 3/2*n*√(n^2-ℓ^2) | |||
|
|
|||
| A = powneg1(n′-ℓ)/(4*factorial(2ℓ-1)) | |||
| B = √(factorial_ratio(n+ℓ,n-ℓ-1)*factorial_ratio(n′+ℓ-1,n′-ℓ)) | |||
| B = √(factorial(n+ℓ,n-ℓ-1)*factorial(n′+ℓ-1,n′-ℓ)) | |||
There was a problem hiding this comment.
Wouldn't the gamma function from SpecialFunctions be better here, for the high n/ell cases?
There was a problem hiding this comment.
I am actually unsure, when talking about ratios of factorials. Should be investigated. This is of course related to #2, however, a numerically stable evaluation of the formula will not help for arbitrarily large values of n/ell, anyway, since the dipole moments (in the length gauge) are analytically divergent. This is physically reasonable, since the matrix element is ~ <n' ell' | r | n ell>, which of course will increase, the larger the values of the quantum numbers are.
Codecov Report
@@ Coverage Diff @@
## master #7 +/- ##
===========================================
- Coverage 32.00% 21.90% -10.10%
===========================================
Files 6 5 -1
Lines 100 105 +5
===========================================
- Hits 32 23 -9
- Misses 68 82 +14
Continue to review full report at Codecov.
|
