Ostrich (#15)

* Ignoring invalid division by zero or nan An ostrich approach to #11 * Dropzweights (#14) * remove full nan slices * don't acr by default * look for max of model only at unflagged locations * drop invalid gausspars * accommodate different keys for different header * idx -> fidx * fresq -> freqs * Exclude any bands that might be awful with `-ds` * pep8 indetation level * syntax error fix * Update spi_fitter.py * Update spi_fitter.py --------- Co-authored-by: landmanbester <[email protected]> * Fix freq dimension check Fix empty gauspar checking --------- Co-authored-by: Athanaseus Javas Ramaila <[email protected]> Co-authored-by: landmanbester <[email protected]>
landmanbester · Apr 10, 2024 · fef25e9 · fef25e9
1 parent a22c45d
commit fef25e9
Show file tree

Hide file tree

Showing 2 changed files with 27 additions and 10 deletions.
diff --git a/spimple/apps/image_convolver.py b/spimple/apps/image_convolver.py
@@ -63,14 +63,18 @@ def image_convolver():
     l_coord -= ref_l
     m_coord, ref_m = data_from_header(hdr, axis=2)
     m_coord -= ref_m
-    if hdr["CTYPE4"].lower() == 'freq':
-        freq_axis = 4
-    elif hdr["CTYPE3"].lower() == 'freq':
-        freq_axis = 3
+
+    if hdr["NAXIS"] >2:
+        if "CTYPE4" in hdr and hdr["CTYPE4"].lower() == 'freq':
+            freq_axis = 4
+        elif hdr["CTYPE3"].lower() == 'freq':
+            freq_axis = 3
+        else:
+            raise ValueError("Freq axis must be 3rd or 4th")
+        freqs, ref_freq = data_from_header(hdr, axis=freq_axis)
     else:
-        raise ValueError("Freq axis must be 3rd or 4th")
-    freqs, ref_freq = data_from_header(hdr, axis=freq_axis)
-
+        freqs = np.array([0])
+
     nchan = freqs.size
     gausspari = ()
     if freqs.size > 1:
@@ -151,6 +155,7 @@ def image_convolver():
     if imagei.ndim != 3:
         raise ValueError("Unsupported number of image dimensions")
     print('Convolving image', file=log)
+
     image, gausskern = convolve2gaussres(imagei, xx, yy, gaussparf,
                                          opts.nthreads, gausspari,
                                          opts.padding_frac)

diff --git a/spimple/utils.py b/spimple/utils.py
@@ -89,8 +89,16 @@ def Gaussian2D(xin, yin, GaussPar=(1., 1., 0.), normalise=True):
     S0, S1, PA = GaussPar
     Smaj = np.maximum(S0, S1)
     Smin = np.minimum(S0, S1)
-    A = np.array([[1. / Smin ** 2, 0],
+
+    # hackaround for runtime warning
+    if 0 in (Smaj, Smin):
+        print("Nan values have been encountered. Subverting RuntimeWarning")
+        A =  np.array([[np.nan, 0],
+                  [0, np.nan]])
+    else:
+        A = np.array([[1. / Smin ** 2, 0],
                   [0, 1. / Smaj ** 2]])
+
 
     c, s, t = np.cos, np.sin, np.deg2rad(-PA)
     R = np.array([[c(t), -s(t)],
@@ -157,15 +165,19 @@ def convolve2gaussres(image, xx, yy, gaussparf, nthreads, gausspari=None, pfrac=
     imhat = r2c(iFs(image, axes=ax), axes=ax, forward=True, nthreads=nthreads, inorm=0)
 
     # convolve to desired resolution
-    if gausspari is None:
+    if gausspari in [None, ()]:
         imhat *= gausskernhat
     else:
         for i in range(nband):
             thiskern = Gaussian2D(xx, yy, gausspari[i], normalise=norm_kernel).astype(np.float64)
             thiskern = np.pad(thiskern[None], padding, mode='constant')
             thiskernhat = r2c(iFs(thiskern, axes=ax), axes=ax, forward=True, nthreads=nthreads, inorm=0)
 
-            convkernhat = np.where(np.abs(thiskernhat)>1e-10, gausskernhat/thiskernhat, 0.0)
+            if not np.all(np.isnan(thiskernhat)):
+                convkernhat = np.where(np.abs(thiskernhat)>1e-10, gausskernhat/thiskernhat, 0.0)
+            else:
+                print("Nan values have been encountered. Subverting RuntimeWarning")
+                convkernhat = np.zeros_like(thiskernhat).astype("complex")
 
             imhat[i] *= convkernhat[0]