implement scipy interpolate? #211

Activity

1. Oceania2018 commented on Mar 12, 2019

   

   Oceania2018

   on Mar 12, 2019

   Member

   More actions

   Will do Interpolate a 1-D function. and 2-D

   👍React with 👍1Reacted by Yucheng Chien
2. Oceania2018 commented on Mar 26, 2019

   

   Oceania2018

   on Mar 26, 2019

   Member

   More actions

   @xinqipony Is https://docs.scipy.org/doc/numpy/reference/generated/numpy.interp.html you want?
3. xinqipony commented on Mar 26, 2019

   

   xinqipony

   on Mar 26, 2019

   Author

   More actions

   yes, this is what I used:
   interpolate.interp1d(x, y, kind='cubic', axis=0),

   in the meantime, I use MathNet now, but its implementation is not so good and not easy to use,

   look forward to your new updates!
4. Oceania2018

   self-assigned this

   on Mar 26, 2019
5. Oceania2018

   added

   enhancementNew feature or requestNew feature or request

   on Mar 26, 2019
6. Nucs commented on Sep 7, 2026

   

   Nucs

   on Sep 7, 2026

   Member

   More actions

   @xinqipony, the NumPy piece shipped in 0.70.0 (#628): np.interp (1-D linear, with left/right/period), bit-exact vs NumPy:

   var xp = np.array(new[] {1.0, 2, 3});
   var fp = np.array(new[] {10.0, 20, 30});
   
   Console.WriteLine(np.interp(np.array(new[] {2.5}), xp, fp));   // [25.]
   Console.WriteLine(np.interp(np.array(new[] {0.5}), xp, fp));   // [10.]  (clamped to fp[0])

   Full SciPy interpolation (splines, griddata) is out of scope for a NumPy port. Closing.
7. Nucs

   closed this as completedon Sep 7, 2026