Auto-tools/Projects/OrangeFactor/Statistics: Difference between revisions

Auto-tools/Projects/OrangeFactor/Statistics (view source)

1,197 bytes added , 2 November 2010

→‎Smoothing: note weighted smoothing

947

edits

@@ Line 51: / Line 51: @@
         retval[i] += data[i] + data[i-1]
         denom += 2.
+     retval[i] /= denom
+   return retval
+This function performs one pass of a smoothing filter, which for an interior node gives the average of two parts the value plus one part the value of each of its neighbors.  Multiple passes of the smoothing function may be applied to the data to achieve even smoother results.  The exact number of passes to use depends on the data to be smoothed -- too few passes and the data will be too rough, too many passes and the data will lose its shape.  It should be noted that smoothing is a lossy operation.
+For each pass of the above function, data propagates one node to the right and one node to the left.  Often, it is desirable that smoothing take place over a wider swath of data but without the comparatively high coefficients given to the neighboring nodes.  In this case, weighted smoothing may be used:
+ def smoothing_pass(data, weight=1.):
+   if len(data) < 2:
+     return data
+   retval = [0 for i in data]
+   for i in range(len(data)):
+     denom = 0.
+     if i < len(data) - 1:
+       retval[i] += data[i] + weight*data[i+1]
+       denom += 1. + weight
+     if i > 0:
+       retval[i] += data[i] + weight*data[i-1]
+       denom += 1. + weight
       retval[i] /= denom
     return retval