v1.0.2 - bug fix and change from deepdish to h5py

bagpipes/fitting/fit.py

@@ -4,7 +4,7 @@
 import os
 import time
 import warnings
-import deepdish as dd
+import h5py
 
 from copy import deepcopy
 
@@ -74,15 +74,21 @@ def __init__(self, galaxy, fit_instructions, run=".", time_calls=False,
         self.fname = "pipes/posterior/" + run + "/" + self.galaxy.ID + "_"
 
         # A dictionary containing properties of the model to be saved.
-        self.results = {"fit_instructions": self.fit_instructions}
+        self.results = {}
 
         # If a posterior file already exists load it.
         if os.path.exists(self.fname[:-1] + ".h5"):
-            self.results = dd.io.load(self.fname[:-1] + ".h5")
+            file = h5py.File(self.fname[:-1] + ".h5", "r")
+
             self.posterior = posterior(self.galaxy, run=run,
                                        n_samples=n_posterior)
-            self.fit_instructions = dd.io.load(self.fname[:-1] + ".h5",
-                                               group="/fit_instructions")
+
+            self.fit_instructions = eval(file.attrs["fit_instructions"])
+
+            for k in file.keys():
+                self.results[k] = np.array(file[k])
+                if np.sum(self.results[k].shape) == 1:
+                    self.results[k] = self.results[k][0]
 
             if rank == 0:
                 print("\nResults loaded from " + self.fname[:-1] + ".h5\n")
@@ -136,18 +142,23 @@ def fit(self, verbose=False, n_live=400, use_MPI=True):
             samples2d = np.loadtxt(self.fname + "post_equal_weights.dat")
             lnz_line = open(self.fname + "stats.dat").readline().split()
 
+            file = h5py.File(self.fname[:-1] + ".h5", "w")
+
+            file.attrs["fit_instructions"] = str(self.fit_instructions)
+
             self.results["samples2d"] = samples2d[:, :-1]
             self.results["lnlike"] = samples2d[:, -1]
             self.results["lnz"] = float(lnz_line[-3])
             self.results["lnz_err"] = float(lnz_line[-1])
             self.results["median"] = np.median(samples2d, axis=0)
             self.results["conf_int"] = np.percentile(self.results["samples2d"],
                                                      (16, 84), axis=0)
+            for k in self.results.keys():
+                file.create_dataset(k, data=self.results[k])
+
+            self.results["fit_instructions"] = self.fit_instructions
 
-            # Save re-formatted outputs as HDF5 and remove MultiNest output.
-            with warnings.catch_warnings():
-                warnings.simplefilter("ignore")
-                dd.io.save(self.fname[:-1] + ".h5", self.results)
+            file.close()
 
             os.system("rm " + self.fname + "*")
 

bagpipes/fitting/posterior.py

@@ -3,7 +3,7 @@
 import numpy as np
 
 import os
-import deepdish as dd
+import h5py
 
 from copy import deepcopy
 
@@ -48,11 +48,13 @@ def __init__(self, galaxy, run=".", n_samples=500):
             raise IOError("Fit results not found for " + self.galaxy.ID + ".")
 
         # Reconstruct the fitted model.
-        self.fit_instructions = dd.io.load(fname, group="/fit_instructions")
+        file = h5py.File(fname, "r")
+
+        self.fit_instructions = eval(file.attrs["fit_instructions"])
         self.fitted_model = fitted_model(self.galaxy, self.fit_instructions)
 
         # 2D array of samples for the fitted parameters only.
-        self.samples2d = dd.io.load(fname, group="/samples2d")
+        self.samples2d = np.array(file["samples2d"])
 
         # If fewer than n_samples exist in posterior, reduce n_samples
         if self.samples2d.shape[0] < self.n_samples:
@@ -128,7 +130,7 @@ def get_basic_quantities(self):
 
         quantity_names = ["stellar_mass", "formed_mass", "sfr", "ssfr", "nsfr",
                           "mass_weighted_age", "tform", "tquench",
-                          "mass_weighted_metallicity"]
+                          "mass_weighted_zmet"]
 
         for q in quantity_names:
             self.samples[q] = np.zeros(self.n_samples)

bagpipes/models/star_formation_history.py

@@ -134,11 +134,11 @@ def _calculate_derived_quantities(self):
         self.mass_weighted_age = np.sum(self.sfh*self.age_widths*self.ages)
         self.mass_weighted_age /= np.sum(self.sfh*self.age_widths)
 
-        self.mass_weighted_met = np.sum(self.live_frac_grid*self.ceh.grid,
+        self.mass_weighted_zmet = np.sum(self.live_frac_grid*self.ceh.grid,
                                         axis=1)
-        self.mass_weighted_met /= np.sum(self.live_frac_grid*self.ceh.grid)
-        self.mass_weighted_met *= config.metallicities
-        self.mass_weighted_met = np.sum(self.mass_weighted_met)
+        self.mass_weighted_zmet /= np.sum(self.live_frac_grid*self.ceh.grid)
+        self.mass_weighted_zmet *= config.metallicities
+        self.mass_weighted_zmet = np.sum(self.mass_weighted_zmet)
 
         self.tform = self.age_of_universe - self.mass_weighted_age
 

bagpipes/plotting/general.py

@@ -255,12 +255,12 @@ def auto_axis_label(ax, y_scale, z_non_zero=True, log_x=False):
     if tex_on:
         if z_non_zero:
             ax.set_ylabel("$\\mathrm{f_{\\lambda}}\\ \\mathrm{/\\ 10^{"
-                          + str(y_scale)
+                          + str(int(y_scale))
                           + "}\\ erg\\ s^{-1}\\ cm^{-2}\\ \\AA^{-1}}$")
 
         else:
             ax.set_ylabel("$\\mathrm{f_{\\lambda}}\\ \\mathrm{/\\ 10^{"
-                          + str(y_scale)
+                          + str(int(y_scale))
                           + "}\\ erg\\ s^{-1}\\ \\AA^{-1}}$")
 
         if log_x:

bagpipes/plotting/plot_galaxy.py

@@ -44,7 +44,7 @@ def plot_galaxy(galaxy, show=True, return_y_scale=False, y_scale_spec=None):
     # Add observed photometry to plot
     if galaxy.photometry_exists and galaxy.spectrum_exists:
         phot_ax = plt.subplot(gs[1, 0])
-        y_scale_phot = add_observed_photometry(galaxy, phot_ax)
+        y_scale_phot = add_observed_photometry(galaxy, phot_ax).astype(float)
         y_scale.append(y_scale_phot)
         axes.append(phot_ax)
 

bagpipes/plotting/plot_spectrum.py

@@ -23,7 +23,7 @@ def add_spectrum(spectrum, ax, x_ticks=None, zorder=4, z_non_zero=True,
         ymax = 1.05*np.nanmax(spectrum[:, 1])
 
     if y_scale is None:
-        y_scale = int(np.log10(ymax))-1
+        y_scale = float(int(np.log10(ymax))-1)
 
     ax.set_ylim(0., ymax*10**-y_scale)
     ax.set_xlim(spectrum[0, 0], spectrum[-1, 0])

bagpipes/plotting/plot_spectrum_posterior.py

@@ -56,7 +56,7 @@ def add_photometry_posterior(fit, ax, zorder=4, y_scale=None, color1=None,
     ymax = 1.05*np.max(upper_lims[mask])
 
     if not y_scale:
-        y_scale = int(np.log10(ymax))-1
+        y_scale = float(int(np.log10(ymax))-1)
 
     # Calculate posterior median redshift.
     if "redshift" in fit.fitted_model.params:
@@ -106,7 +106,7 @@ def add_spectrum_posterior(fit, ax, zorder=4, y_scale=None):
     ymax = 1.05*np.max(fit.galaxy.spectrum[:, 1])
 
     if not y_scale:
-        y_scale = int(np.log10(ymax))-1
+        y_scale = float(int(np.log10(ymax))-1)
 
     wavs = fit.galaxy.spectrum[:, 0]
     spec_post = np.copy(fit.posterior.samples["spectrum"])

docs/conf.py

@@ -54,9 +54,9 @@
 # built documents.
 #
 # The short X.Y version.
-version = u'1.0.1'
+version = u'1.0.2'
 # The full version, including alpha/beta/rc tags.
-release = u'1.0.1'
+release = u'1.0.2'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

rtd-environment.yml

@@ -5,12 +5,12 @@ dependencies:
     - astropy
     - scipy
     - matplotlib>=2.2.2
-    - numpy>=1.14.2
+    - numpy
     - pybind11
+    - h5py
     - pip:
       - corner
       - pymultinest
       - msgpack
-      - deepdish
-      - pandas<=1.1.5
+      - pandas
       - spectres

setup.py

@@ -11,7 +11,7 @@
 setup(
     name='bagpipes',
 
-    version='1.0.1',
+    version='1.0.2',
 
     description='Galaxy spectral fitting',
 
@@ -29,9 +29,9 @@
 
     include_package_data=True,
 
-    install_requires=["numpy>=1.14.2", "corner", "pymultinest>=2.11",
+    install_requires=["numpy", "corner", "pymultinest>=2.11", "h5py", "pandas",
                       "astropy", "matplotlib>=2.2.2", "scipy", "msgpack",
-                      "deepdish", "pandas<=1.1.5", "spectres"],
+                      "spectres"],
 
     project_urls={
         "readthedocs": "https://bagpipes.readthedocs.io",