Is there a library module or other straightforward way to implement multivariate spline interpolation in python?
Specifically, I have a set of scalar data on a regularly-spaced three-dimensional grid which I need to interpolate at a small number of points scattered throughout the domain. For two dimensions, I have been using scipy.interpolate.RectBivariateSpline, and I'm essentially looking for an extension of that to three-dimensional data.
The N-dimensional interpolation routines I have found are not quite good enough: I would prefer splines over LinearNDInterpolator for smoothness, and I have far too many data points (often over one million) for, e.g., a radial basis function to work.
If anyone knows of a python library that can do this, or perhaps one in another language that I could call or port, I'd really appreciate it.
If I'm understanding your question correctly, your input "observation" data is regularly gridded?
If so, scipy.ndimage.map_coordinates does exactly what you want.
It's a bit hard to understand at first pass, but essentially, you just feed it a sequence of coordinates that you want to interpolate the values of the grid at in pixel/voxel/n-dimensional-index coordinates.
As a 2D example:
import numpy as np
from scipy import ndimage
import matplotlib.pyplot as plt
# Note that the output interpolated coords will be the same dtype as your input
# data. If we have an array of ints, and we want floating point precision in
# the output interpolated points, we need to cast the array as floats
data = np.arange(40).reshape((8,5)).astype(np.float)
# I'm writing these as row, column pairs for clarity...
coords = np.array([[1.2, 3.5], [6.7, 2.5], [7.9, 3.5], [3.5, 3.5]])
# However, map_coordinates expects the transpose of this
coords = coords.T
# The "mode" kwarg here just controls how the boundaries are treated
# mode='nearest' is _not_ nearest neighbor interpolation, it just uses the
# value of the nearest cell if the point lies outside the grid. The default is
# to treat the values outside the grid as zero, which can cause some edge
# effects if you're interpolating points near the edge
# The "order" kwarg controls the order of the splines used. The default is
# cubic splines, order=3
zi = ndimage.map_coordinates(data, coords, order=3, mode='nearest')
row, column = coords
nrows, ncols = data.shape
im = plt.imshow(data, interpolation='nearest', extent=[0, ncols, nrows, 0])
plt.colorbar(im)
plt.scatter(column, row, c=zi, vmin=data.min(), vmax=data.max())
for r, c, z in zip(row, column, zi):
plt.annotate('%0.3f' % z, (c,r), xytext=(-10,10), textcoords='offset points',
arrowprops=dict(arrowstyle='->'), ha='right')
plt.show()
To do this in n-dimensions, we just need to pass in the appropriate sized arrays:
import numpy as np
from scipy import ndimage
data = np.arange(3*5*9).reshape((3,5,9)).astype(np.float)
coords = np.array([[1.2, 3.5, 7.8], [0.5, 0.5, 6.8]])
zi = ndimage.map_coordinates(data, coords.T)
As far as scaling and memory usage goes, map_coordinates will create a filtered copy of the array if you're using an order > 1 (i.e. not linear interpolation). If you just want to interpolate at a very small number of points, this is a rather large overhead. It doesn't increase with the number points you want to interpolate at, however. As long as have enough RAM for a single temporary copy of your input data array, you'll be fine.
If you can't store a copy of your data in memory, you can either a) specify prefilter=False and order=1 and use linear interpolation, or b) replace your original data with a filtered version using ndimage.spline_filter, and then call map_coordinates with prefilter=False.
Even if you have enough ram, keeping the filtered dataset around can be a big speedup if you need to call map_coordinates multiple times (e.g. interactive use, etc).
Smooth spline interpolation in dim > 2 is difficult to implement, and so there are not many freely available libraries able to do that (in fact, I don't know any).
You can try inverse distance weighted interpolation, see: Inverse Distance Weighted (IDW) Interpolation with Python .
This should produce reasonably smooth results, and scale better than RBF to larger data sets.
Related
The task at hand is seemingly simple:
I have a 2D grid of data. The data is available in 2D arrays for X and Y coordinates, as well as the input variable which I want to interpolate. This means I can plot the data using rectangular cells, which means it is possible to use bilinear interpolation. Unfortunately, the data is not precisely aligned with the coordinates, and also not precisely spaced. There were some numerics involved in creating the data, which means that all sampling locations are a little off the mark, and the cell spacing is smooth but not uniform.
I would like to interpolate from this input grid to a set of predefined sample coordinates (as opposed to simply refining the mesh).
In short, an example for my type of input is:
# a nice, regular grid
Xs, Ys = np.meshgrid(np.linspace(0, 1, num=3), np.linspace(0, 1, num=5))
# ...perturbed by some systematic and some random noise ...
X_in = Xs + np.random.normal(scale=0.03, size=(5, 3))
# ...and some systematic deviation
Y_in = (Ys + np.random.normal(scale=0.03, size=(5, 3)))* (1 + Xs**1.5)
# and some variable at each node to interpolate
Z_in = np.random.normal(scale=1, size=(5, 3))
So (X_in, Y_in) are arrays of shape (n, m) which define a mesh with quadrilateral cells, and Z_in another array of ther same shape which provides a value at each node in that mesh. I am looking for some Python library that performs bilinear interpolation of Z_in across those cells.
However, all methods I have found so far either ignore the rectangular structure (and triangulate the data, or fit some 2D spline through arbitrary point clouds), or require a perfectly rectangular and equally-spaced grid as input (which mine is not).
Examples of answers/methods that seem not to be appliccable:
This answer recommends using scipy.ndimage.map_coordinates -- but that effectively uses the indices of the 2D input data array as coordinates, which won't work for me.
scipy.interpolate.interp2d requires either a regular grid (node locations provided by 1D X and Y arrays), or an irregular one, which is flattened, which means that the algorithm cannot know which nodes form a cell. This means it either fits some spline through unstructured data, or triangulates it. And it only interpolates onto regular grids or individual points.
scipy.interpolate.RectBivariateSpline is recommended for interpolation from gridded data but only accepts input points which are perfectly aligned with the coordinate system.
There's also a Matplotlib toolkit for interpolation, which I had thought should be able to do this sort of thing, as it also does interpolated contour plots of rectangular meshes, but as it turns out, even though mpl_toolkits.basemap.interp accepts arbitrary quadrilateral meshes as target for interpolation, it cannot use them as inputs ...
Upon closer inspection, it turns out that even matplotlib.plt.contour() does not seem to perform bilinear interpolation when plotting the input data:
plt.contour(X_in, Y_in, Z_in, levels=np.linspace(Z_in.min(), Z_in.max(), 50))
plt.plot(X_in, Y_in, 'k-')
plt.plot(X_in.T, Y_in.T, 'k-')
As you can see, the contour lines within the cells are straight, but with bilinear interpolation, they should not be, and there should not be those empty quadrilateral areas in the mittle of some cells. I suspect that Matplotlib only finds the contour values on the cell edges and simply draws straight lines between them.
I have found two explanations of the maths of bilinear interpolation from grids which are not perfectly aligned, but I was hoping to come across a ready-made implementation somewhere because I'm sure that this kind of task is not so rare, and a numpy or scipy implementation (if it exists) is probably way faster than whatever I'd implement myself.
So, I have three numpy arrays which store latitude, longitude, and some property value on a grid -- that is, I have LAT(y,x), LON(y,x), and, say temperature T(y,x), for some limits of x and y. The grid isn't necessarily regular -- in fact, it's tripolar.
I then want to interpolate these property (temperature) values onto a bunch of different lat/lon points (stored as lat1(t), lon1(t), for about 10,000 t...) which do not fall on the actual grid points. I've tried matplotlib.mlab.griddata, but that takes far too long (it's not really designed for what I'm doing, after all). I've also tried scipy.interpolate.interp2d, but I get a MemoryError (my grids are about 400x400).
Is there any sort of slick, preferably fast way of doing this? I can't help but think the answer is something obvious... Thanks!!
Try the combination of inverse-distance weighting and
scipy.spatial.KDTree
described in SO
inverse-distance-weighted-idw-interpolation-with-python.
Kd-trees
work nicely in 2d 3d ..., inverse-distance weighting is smooth and local,
and the k= number of nearest neighbours can be varied to tradeoff speed / accuracy.
There is a nice inverse distance example by Roger Veciana i Rovira along with some code using GDAL to write to geotiff if you're into that.
This is of coarse to a regular grid, but assuming you project the data first to a pixel grid with pyproj or something, all the while being careful what projection is used for your data.
A copy of his algorithm and example script:
from math import pow
from math import sqrt
import numpy as np
import matplotlib.pyplot as plt
def pointValue(x,y,power,smoothing,xv,yv,values):
nominator=0
denominator=0
for i in range(0,len(values)):
dist = sqrt((x-xv[i])*(x-xv[i])+(y-yv[i])*(y-yv[i])+smoothing*smoothing);
#If the point is really close to one of the data points, return the data point value to avoid singularities
if(dist<0.0000000001):
return values[i]
nominator=nominator+(values[i]/pow(dist,power))
denominator=denominator+(1/pow(dist,power))
#Return NODATA if the denominator is zero
if denominator > 0:
value = nominator/denominator
else:
value = -9999
return value
def invDist(xv,yv,values,xsize=100,ysize=100,power=2,smoothing=0):
valuesGrid = np.zeros((ysize,xsize))
for x in range(0,xsize):
for y in range(0,ysize):
valuesGrid[y][x] = pointValue(x,y,power,smoothing,xv,yv,values)
return valuesGrid
if __name__ == "__main__":
power=1
smoothing=20
#Creating some data, with each coodinate and the values stored in separated lists
xv = [10,60,40,70,10,50,20,70,30,60]
yv = [10,20,30,30,40,50,60,70,80,90]
values = [1,2,2,3,4,6,7,7,8,10]
#Creating the output grid (100x100, in the example)
ti = np.linspace(0, 100, 100)
XI, YI = np.meshgrid(ti, ti)
#Creating the interpolation function and populating the output matrix value
ZI = invDist(xv,yv,values,100,100,power,smoothing)
# Plotting the result
n = plt.normalize(0.0, 100.0)
plt.subplot(1, 1, 1)
plt.pcolor(XI, YI, ZI)
plt.scatter(xv, yv, 100, values)
plt.title('Inv dist interpolation - power: ' + str(power) + ' smoothing: ' + str(smoothing))
plt.xlim(0, 100)
plt.ylim(0, 100)
plt.colorbar()
plt.show()
There's a bunch of options here, which one is best will depend on your data...
However I don't know of an out-of-the-box solution for you
You say your input data is from tripolar data. There are three main cases for how this data could be structured.
Sampled from a 3d grid in tripolar space, projected back to 2d LAT, LON data.
Sampled from a 2d grid in tripolar space, projected into 2d LAT LON data.
Unstructured data in tripolar space projected into 2d LAT LON data
The easiest of these is 2. Instead of interpolating in LAT LON space, "just" transform your point back into the source space and interpolate there.
Another option that works for 1 and 2 is to search for the cells that maps from tripolar space to cover your sample point. (You can use a BSP or grid type structure to speed up this search) Pick one of the cells, and interpolate inside it.
Finally there's a heap of unstructured interpolation options .. but they tend to be slow.
A personal favourite of mine is to use a linear interpolation of the nearest N points, finding those N points can again be done with gridding or a BSP. Another good option is to Delauney triangulate the unstructured points and interpolate on the resulting triangular mesh.
Personally if my mesh was case 1, I'd use an unstructured strategy as I'd be worried about having to handle searching through cells with overlapping projections. Choosing the "right" cell would be difficult.
I suggest you taking a look at GRASS (an open source GIS package) interpolation features (http://grass.ibiblio.org/gdp/html_grass62/v.surf.bspline.html). It's not in python but you can reimplement it or interface with C code.
Am I right in thinking your data grids look something like this (red is the old data, blue is the new interpolated data)?
alt text http://www.geekops.co.uk/photos/0000-00-02%20%28Forum%20images%29/DataSeparation.png
This might be a slightly brute-force-ish approach, but what about rendering your existing data as a bitmap (opengl will do simple interpolation of colours for you with the right options configured and you could render the data as triangles which should be fairly fast). You could then sample pixels at the locations of the new points.
Alternatively, you could sort your first set of points spatially and then find the closest old points surrounding your new point and interpolate based on the distances to those points.
There is a FORTRAN library called BIVAR, which is very suitable for this problem. With a few modifications you can make it usable in python using f2py.
From the description:
BIVAR is a FORTRAN90 library which interpolates scattered bivariate data, by Hiroshi Akima.
BIVAR accepts a set of (X,Y) data points scattered in 2D, with associated Z data values, and is able to construct a smooth interpolation function Z(X,Y), which agrees with the given data, and can be evaluated at other points in the plane.
hi i have two sets of data taken from two seperate import files which are both being imported into python and have been placed in two seperate lists as follows:
list 1 is of the form:
(node, x coordinate, y coordinate, z coordinate)
example list 1: [[1,0,0,0],[2,1,0,0],[3,0,1,0],[4,1,1,0],[5,0,0,1],[6,1,0,1],[7,0,1,1],[8,1,1,1]]
list 2 is in the form:
(x coordinate, y coordinate, z coordinate, temperature)
example list 2: [[0,0,0,100],[1,0,0,90],[0,1,0,85],[1,1,0,110],[0,0,1,115],[1,0,1,118],[0,1,1,100],[1,1,11,96]]
from these two lists I need to use the coordinates to create a third list which contains a node value and its corresponding temperature. This task is a simple dictionary function if all the x y and z coordinates match up however with the data i am working with this will not always be the case.
For example if in list 1 I add a new entry at the end of the list, node number 9;
new entry at end of list 1 [9, 0.5, 0.9, 0.25]
Now I find myself with a node number with no corresponding temperature. At this point an interpolation function will need to be performed on list 2 to give me the temperature related to this node. Through basic 3d interpolation calculations I have worked out that this temperature will be 97.9 therefore my final output list would look like this:
Output list:
(node, temperature)
Output list: [[1,100],[2,90],[3,85],[4,110],[5,115],[6,118],[7,100],[8,96],[9,97.9]]
I am reasonably new to python so am struggling to find a solution to this interpolation problem, I have been researching how to do this for a number of weeks now and have still not been able to find a solution.
Any help would be greatly greatly appreciated,
Thanks
There are quite a few interpolation routines in scipy, but above 2 dimensions, most of them only offer linear and nearest neighbour interpolation - which might not be sufficient for your use.
All of the interpolation routiens are listed on the interplation page of the scipy docs area. Straight away you can ignore the mnivariate, and 1D and 2D spline sections - you want the multivariate section.
There are 9 functions here, split into structured and unstructed data:
Unstructured data:
griddata(points, values, xi[, method, ...]) Interpolate unstructured
D-dimensional data.
LinearNDInterpolator(points, values[, ...]) Piecewise linear interpolant in N dimensions.
NearestNDInterpolator(points, values) Nearest-neighbour interpolation in N dimensions.
CloughTocher2DInterpolator(points, values[, tol]) Piecewise cubic, C1 smooth, curvature-minimizing interpolant in 2D.
Rbf(*args) A class for radial basis function approximation/interpolation of n-dimensional scattered data.
interp2d(x, y, z[, kind, copy, ...]) Interpolate over a 2-D grid. For >
data on a grid:
interpn(points, values, xi[, method, ...]) Multidimensional
interpolation on regular grids.
RegularGridInterpolator(points, values[, ...]) Interpolation on a regular grid in arbitrary dimensions
RectBivariateSpline(x, y, z[, bbox, kx, ky, s]) Bivariate spline approximation over a rectangular mesh.
plus an additional one in the see also section, though we'll ignore that.
You should read how they each work, it might help you understand a little better.
The way these functions work though, is that you pass them data i.e. x,y,z coords, and the corresponding values at those points, and they then return a function which allows you to get a point at any location.
I would recommend the Rbf function here though, as from what i can see it's the only nD option which does not limit you to linear or nearest neighbour interpolation.
For example, you have two lists:
node_locations = [(node, x_coord, y_coord, z_coord), ...]
temp_data = [(x0, y0, z0, temp0), (x1, y1, z1, temp1), ...]
xs, ys, zs, temps = zip(*teemp_data) # This will unpack your data into columns, rather than rows.
from scipy.interpolate import Rbf
rbfi = Rbf(xs, ys, zs, temps)
# I don't know how you want your output data, so i'm just dumping it in a dictionary.
node_data = {}
for node, x, y, z in node_locations:
node_data[node] = rbfi(x, y, z)
Try something like that.
For scientific computing, I wouldn't use lists but numpy arrays instead.
So in your case:
import numpy as np
nodes = np.array(example_list_1)
temperatures = np.array(example_list_2)
With this you can then go on to use scipy's interpolation functions, like for example:
http://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.griddata.html#scipy.interpolate.griddata
from scipy.interpolate import griddata
interpolated = griddata(temperatures[:, :-1],
temperatures[:, -1],
nodes[:, 1:])
I have some data over a 2D range that I am interested in analyzing. These data were originally in lists x,y, and z where z[i] was the value for the point located at (x[i],y[i]). I then interpolated this data onto a regular grid using
x=np.array(x)
y=np.array(y)
z=np.array(z)
xi=np.linspace(minx,maxx,100)
yi=np.linspace(miny,maxy,100)
zi=griddata(x,y,z,xi,yi)
I then plotted the xi,yi,zi data using
plt.contour(xi,yi,zi)
plt.pcolormesh(xi,yi,zi,cmap=plt.get_cmap('PRGn'),norm=plt.Normalize(-10,10),vmin=-10,vmax=10)
This produced this plot:
In this plot you can see the S-like curve where the values are equal to zero (aside: the data doesn't vary as rapidly as shown in the colorbar -- that's simply a result of me normalizing the data to -10-10 when it actually extends far beyond that range; I did this to make the zero-valued region show up better -- maybe there's a better way of doing this too...).
The scattered dots are simply the points at which I have original data (yes, in this case my data was already on a regular grid). What I'm curious about is whether there is a good way for me to extract the values for which the curve is zero and obtain x,y pairs that, if plotted as a line, would trace that zero-region in the colormesh. I could interpolate to a really fine grid and then just brute force search for the values which are closest to zero. But is there a more automatic way of doing this, or a more automatic way of plotting this "zero-line"?
And a secondary question: I am using griddata correctly, right? I have these simple 1D arrays although elsewhere people use various meshgrids, loading texts, etc., before calling griddata.
Here is a full example:
import numpy as np
import matplotlib.pyplot as plt
y, x = np.ogrid[-1.5:1.5:200j, -1.5:1.5:200j]
f = (x**2 + y**2)**4 - (x**2 - y**2)**2
plt.figure(figsize=(9,4))
plt.subplot(121)
extent = [np.min(x), np.max(x), np.min(y), np.max(y)]
cs = plt.contour(f, extent=extent, levels=[0.1],
colors=["b", "r"], linestyles=["solid", "dashed"], linewidths=[2, 2])
plt.subplot(122)
# get the points on the lines
for c in cs.collections:
data = c.get_paths()[0].vertices
plt.plot(data[:,0], data[:,1],
color=c.get_color()[0], linewidth=c.get_linewidth()[0])
plt.show()
here is the output:
I need to interpolate temperature data linearly in 4 dimensions (latitude, longitude, altitude and time).
The number of points is fairly high (360x720x50x8) and I need a fast method of computing the temperature at any point in space and time within the data bounds.
I have tried using scipy.interpolate.LinearNDInterpolator but using Qhull for triangulation is inefficient on a rectangular grid and takes hours to complete.
By reading this SciPy ticket, the solution seemed to be implementing a new nd interpolator using the standard interp1d to calculate a higher number of data points, and then use a "nearest neighbor" approach with the new dataset.
This, however, takes a long time again (minutes).
Is there a quick way of interpolating data on a rectangular grid in 4 dimensions without it taking minutes to accomplish?
I thought of using interp1d 4 times without calculating a higher density of points, but leaving it for the user to call with the coordinates, but I can't get my head around how to do this.
Otherwise would writing my own 4D interpolator specific to my needs be an option here?
Here's the code I've been using to test this:
Using scipy.interpolate.LinearNDInterpolator:
import numpy as np
from scipy.interpolate import LinearNDInterpolator
lats = np.arange(-90,90.5,0.5)
lons = np.arange(-180,180,0.5)
alts = np.arange(1,1000,21.717)
time = np.arange(8)
data = np.random.rand(len(lats)*len(lons)*len(alts)*len(time)).reshape((len(lats),len(lons),len(alts),len(time)))
coords = np.zeros((len(lats),len(lons),len(alts),len(time),4))
coords[...,0] = lats.reshape((len(lats),1,1,1))
coords[...,1] = lons.reshape((1,len(lons),1,1))
coords[...,2] = alts.reshape((1,1,len(alts),1))
coords[...,3] = time.reshape((1,1,1,len(time)))
coords = coords.reshape((data.size,4))
interpolatedData = LinearNDInterpolator(coords,data)
Using scipy.interpolate.interp1d:
import numpy as np
from scipy.interpolate import LinearNDInterpolator
lats = np.arange(-90,90.5,0.5)
lons = np.arange(-180,180,0.5)
alts = np.arange(1,1000,21.717)
time = np.arange(8)
data = np.random.rand(len(lats)*len(lons)*len(alts)*len(time)).reshape((len(lats),len(lons),len(alts),len(time)))
interpolatedData = np.array([None, None, None, None])
interpolatedData[0] = interp1d(lats,data,axis=0)
interpolatedData[1] = interp1d(lons,data,axis=1)
interpolatedData[2] = interp1d(alts,data,axis=2)
interpolatedData[3] = interp1d(time,data,axis=3)
Thank you very much for your help!
In the same ticket you have linked, there is an example implementation of what they call tensor product interpolation, showing the proper way to nest recursive calls to interp1d. This is equivalent to quadrilinear interpolation if you choose the default kind='linear' parameter for your interp1d's.
While this may be good enough, this is not linear interpolation, and there will be higher order terms in the interpolation function, as this image from the wikipedia entry on bilinear interpolation shows:
This may very well be good enough for what you are after, but there are applications where a triangulated, really piecewise linear, interpoaltion is preferred. If you really need this, there is an easy way of working around the slowness of qhull.
Once LinearNDInterpolator has been setup, there are two steps to coming up with an interpolated value for a given point:
figure out inside which triangle (4D hypertetrahedron in your case) the point is, and
interpolate using the barycentric coordinates of the point relative to the vertices as weights.
You probably do not want to mess with barycentric coordinates, so better leave that to LinearNDInterpolator. But you do know some things about the triangulation. Mostly that, because you have a regular grid, within each hypercube the triangulation is going to be the same. So to interpolate a single value, you could first determine in which subcube your point is, build a LinearNDInterpolator with the 16 vertices of that cube, and use it to interpolate your value:
from itertools import product
def interpolator(coords, data, point) :
dims = len(point)
indices = []
sub_coords = []
for j in xrange(dims) :
idx = np.digitize([point[j]], coords[j])[0]
indices += [[idx - 1, idx]]
sub_coords += [coords[j][indices[-1]]]
indices = np.array([j for j in product(*indices)])
sub_coords = np.array([j for j in product(*sub_coords)])
sub_data = data[list(np.swapaxes(indices, 0, 1))]
li = LinearNDInterpolator(sub_coords, sub_data)
return li([point])[0]
>>> point = np.array([12.3,-4.2, 500.5, 2.5])
>>> interpolator((lats, lons, alts, time), data, point)
0.386082399091
This cannot work on vectorized data, since that would require storing a LinearNDInterpolator for every possible subcube, and even though it probably would be faster than triangulating the whole thing, it would still be very slow.
scipy.ndimage.map_coordinates
is a nice fast interpolator for uniform grids (all boxes the same size).
See multivariate-spline-interpolation-in-python-scipy on SO
for a clear description.
For non-uniform rectangular grids, a simple wrapper
Intergrid maps / scales non-uniform to uniform grids,
then does map_coordinates.
On a 4d test case like yours it takes about 1 μsec per query:
Intergrid: 1000000 points in a (361, 720, 47, 8) grid took 652 msec
For very similar things I use Scientific.Functions.Interpolation.InterpolatingFunction.
import numpy as np
from Scientific.Functions.Interpolation import InterpolatingFunction
lats = np.arange(-90,90.5,0.5)
lons = np.arange(-180,180,0.5)
alts = np.arange(1,1000,21.717)
time = np.arange(8)
data = np.random.rand(len(lats)*len(lons)*len(alts)*len(time)).reshape((len(lats),len(lons),len(alts),len(time)))
axes = (lats, lons, alts, time)
f = InterpolatingFunction(axes, data)
You can now leave it to the user to call the InterpolatingFunction with coordinates:
>>> f(0,0,10,3)
0.7085675631375401
InterpolatingFunction has nice additional features, such as integration and slicing.
However, I do not know for sure whether the interpolation is linear. You would have to look in the module source to find out.
I can not open this address, and find enough informations about this package