nested-fit 5.6.8.dev127__tar.gz → 5.6.10__tar.gz

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nested_fit
-Version: 5.6.8.dev127
+Version: 5.6.10
 Summary: The nested_fit package.
 Author: Martino Trassinelli, Lune Maillard, César Godinho
 Project-URL: Homepage, https://github.com/martinit18/nested_fit

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/nested_fit.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nested_fit
-Version: 5.6.8.dev127
+Version: 5.6.10
 Summary: The nested_fit package.
 Author: Martino Trassinelli, Lune Maillard, César Godinho
 Project-URL: Homepage, https://github.com/martinit18/nested_fit

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/pynested_fit/nested_res.py

@@ -570,7 +570,7 @@ class Analysis(object):
             width = depth = 1
             # Plot
             ax_d = fig_d.add_subplot(111, projection='3d')
-            cmap = plt.cm.get_cmap('jet') # Get desired colormap - you can change this!
+            cmap = plt.get_cmap('jet') # Get desired colormap - you can change this!
             max_height = np.max(zzz)   # get range of colorbars so we can normalize
             min_height = 0.
             # scale each z to [0,1], and get their rgb values
@@ -604,7 +604,7 @@ class Analysis(object):
             width = depth = 1
             # Plot
             ax_f = fig_f.add_subplot(111, projection='3d')
-            cmap = plt.cm.get_cmap('jet') # Get desired colormap - you can change this!
+            cmap = plt.get_cmap('jet') # Get desired colormap - you can change this!
             max_height = np.max(zzz)   # get range of colorbars so we can normalize
             min_height = 0.
             # scale each z to [0,1], and get their rgb values
@@ -638,7 +638,7 @@ class Analysis(object):
             width = depth = 1
             # Plot
             ax_r = fig_r.add_subplot(111, projection='3d')
-            #cmap = plt.cm.get_cmap('jet') # Get desired colormap - you can change this!
+            #cmap = plt.get_cmap('jet') # Get desired colormap - you can change this!
             #max_height = np.max(zzz)   # get range of colorbars so we can normalize
             #min_height = 0.
             # scale each z to [0,1], and get their rgb values
@@ -893,7 +893,7 @@ class Analysis(object):
         plt.xlim(0,shape(data)[0]*1.1)
         plt.xlabel('Sort number')
         plt.ylabel('Value of parameter ' + par_name)
-        cmap=plt.cm.get_cmap('jet')
+        cmap=plt.get_cmap('jet')
         plt.scatter(ix,data[:,par_index],c=data[:,0],linewidth=0.,cmap=cmap)
         cbar = plt.colorbar()
         cbar.set_label('Weight')
@@ -948,7 +948,7 @@ class Analysis(object):
         plt.xlabel(par_name1)
         plt.ylabel(par_name2)
         #plt.ylabel('Value of parameter ' + title)
-        #cmap=plt.cm.get_cmap('jet')
+        #cmap=plt.get_cmap('jet')
         for i in range(ncl): plt.plot(data[data[:,0]==i][:,par_index1],data[data[:,0]==i][:,par_index2],'o')
         plt.tight_layout()
 
@@ -1047,7 +1047,7 @@ class Analysis(object):
         plt.xlabel(par_name1)
         plt.ylabel(par_name2)
         #plt.ylabel('Value of parameter ' + title)
-        #cmap=plt.cm.get_cmap('jet')
+        #cmap=plt.get_cmap('jet')
         plt.plot(data[:,par_index1],data[:,par_index2],'ob')
         plt.tight_layout()
 
@@ -1114,11 +1114,11 @@ class Analysis(object):
 
         # Choose the cmap for likelihood plots
         if cmap == 'normal':
-            cmap=plt.cm.get_cmap('jet')
+            cmap=plt.get_cmap('jet')
         elif cmap == 'accurate':
-            cmap=plt.cm.get_cmap('gist_ncar')
+            cmap=plt.get_cmap('gist_ncar')
         elif cmap == 'moreaccurate':
-            cmap=plt.cm.get_cmap('prism')
+            cmap=plt.get_cmap('prism')
 
         # Plot histogram
         plt.clf()
@@ -1304,11 +1304,11 @@ class Analysis(object):
 
         # Choose the cmap for likelihood plots
         if cmap == 'normal':
-            cmap=plt.cm.get_cmap('jet')
+            cmap=plt.get_cmap('jet')
         elif cmap == 'accurate':
-            cmap=plt.cm.get_cmap('spectral')
+            cmap=plt.get_cmap('nipy_spectral')
         elif cmap == 'moreaccurate':
-            cmap=plt.cm.get_cmap('prism')
+            cmap=plt.get_cmap('prism')
 
         # Plot interpolation
         if clear: plt.clf()

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "nested_fit"
-version = "5.6.8.dev127"
+version = "5.6.10"
 description = "The nested_fit package."
 readme = "README.md"
 
@@ -65,7 +65,7 @@ archs = ["x86_64", "arm64"]
 archs = ["x86_64"]
 
 [tool.setuptools_scm]
-fallback_version = "5.6.8.dev127"
+fallback_version = "1.0.0"
 
 [options]
 python_requires = ">=3.8"

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/src/nested_fit.f90

@@ -560,7 +560,7 @@ PROGRAM NESTED_FIT
          !CALL NESTED_SAMPLING(itry,maxstep_try,nall_try(itry),evsum_final_try(1), &
          !   live_like_final_try(:,1),live_birth_final_try(:,1),live_rank_final_try(:,1),weight_try(:,1),&
          !   live_final_try(:,:,1),live_like_max_try(itry),live_max_try(:,itry), 0, 0)
-         CALL NESTED_SAMPLING(itry,maxstep_try,nall_try(itry),evsum_final_try(1), &
+         CALL NESTED_SAMPLING(itry,maxstep_try,nall_try(itry),evsum_final_try(itry), &
             live_like_max_try(itry),live_max_try(:,itry), 0, 0)
       ELSE
          !CALL NESTED_SAMPLING(itry,maxstep_try,nall_try(itry),evsum_final_try(itry), &

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/src/nested_sampling.f90

@@ -79,6 +79,7 @@ SUBROUTINE NESTED_SAMPLING(itry,maxstep,nall,evsum_final,live_like_max,live_max,
   REAL(8), DIMENSION(4) :: en_write                   ! Vector for writing in energy file  
   ! Rest
   INTEGER(4) :: i, j, l, n, jlim, it
+  INTEGER(4) :: jlow, jhigh, jmid   ! binary search bounds for the insertion point
   REAL(8) :: ADDLOG, rn, gval
   REAL(8) :: moving_eff_avg = 0.
   CHARACTER :: info_string*4096
@@ -363,12 +364,27 @@ SUBROUTINE NESTED_SAMPLING(itry,maxstep,nall,evsum_final,live_like_max,live_max,
         IF (live_like_new(it).GT.live_like(nlive)) THEN
             jlim = nlive
         ELSE
-            DO j=1,nlive-1
-              IF (live_like_new(it).GT.live_like(j).AND.live_like_new(it).LE.live_like(j+1)) THEN
-                 jlim = j
-                 EXIT
-              END IF
+            ! live_like is kept sorted in ascending order, so the insertion point can be
+            ! found by bisection instead of a linear scan. This is O(log nlive) rather
+            ! than O(nlive) and matters a lot for large nlive, where this loop runs once
+            ! per accepted point and averages ~nlive/2 iterations.
+            !
+            ! Invariant: live_like(jlow) < live_like_new(it) <= live_like(jhigh).
+            ! It holds initially because live_like_new(it) > min_live_like = live_like(1)
+            ! (checked above) and live_like_new(it) <= live_like(nlive) in this branch.
+            ! On exit jhigh = jlow + 1, so jlim = jlow is the same index the linear scan
+            ! returned, including when live_like contains repeated values.
+            jlow  = 1
+            jhigh = nlive
+            DO WHILE (jhigh - jlow .GT. 1)
+               jmid = (jlow + jhigh)/2
+               IF (live_like(jmid).LT.live_like_new(it)) THEN
+                  jlow = jmid
+               ELSE
+                  jhigh = jmid
+               END IF
             END DO
+            jlim = jlow
          END IF
          
         ! Store old values

{nested_fit-5.6.8.dev127 → nested_fit-5.6.10}/src/funceval/latex_parser.cpp

@@ -15,6 +15,7 @@
 
 // Lets keep C++11 compliant
 #include <regex>
+#include <functional>
 #include <string>
 #include <cstring>
 #include <iostream>
@@ -23,7 +24,6 @@
 #include <cassert>
 #include <fstream>
 #include <sstream>
-#include <functional>
 
 #ifdef PPROF
 #include <public/tracy/Tracy.hpp>