Save jackknife output in HDF5 file

jackknife.py

 import numpy as np
+import h5py
 
 
+# TODO: Update docstring
+# TODO: Rewrite the docstring with argument descriptions
 def do_jackknife_estimation(input_sample_generator, n_samples,
-                            transformation_function):
-    """Do a jackknife estimation for the transformed sample.
+                            transformation_function,
+                            output_file_name="jackknife.hdf5"):
+    """Do a jackknife estimation for the given samples and function.
 
     Estimate and return the sample mean, variance and standard deviation
-    of the transformed sample y = f(x), where x is the input sample and
+    of the transformed value y = f(x), where x is the input sample and
     f() is the transformation function. The expectation value <y> is
     estimated using the Jackknife algorithm, which does a resampling of
     the input x and then a bias-corrected transformation. For more
@@ -48,10 +52,18 @@ def do_jackknife_estimation(input_sample_generator, n_samples,
     output_variance = sum_of_squares - np.abs(sum_)**2 / n_samples
     output_variance /= (n_samples - 1)
     output_standard_deviation = np.lib.scimath.sqrt(output_variance)
-    return (output_mean, output_variance, output_standard_deviation,
-            transformed_mean)
-
-
+    
+    out_file = h5py.File(output_file_name, "w")
+    out_file.create_dataset("estimated_mean", data=output_mean)
+    out_file.create_dataset("estimated_variance", data=output_variance)
+    out_file.create_dataset("estimated_standard_deviation",
+                            data=output_standard_deviation)
+    out_file.create_dataset("transformed_input_mean", data=transformed_mean)
+    out_file.close()
+    return
+
+
+# TODO: Rewrite the docstring with argument descriptions
 def resample_and_transform(sample, n_samples, sample_mean,
                            transformation_function, transformed_mean):
     """Do a resampling and a bias-corrected transformation.

main.py

@@ -5,6 +5,7 @@ import numpy as np
 
 
 # %%
+# TODO: Make the paths and files command line options
 abinitio_dga = adga.Adga("/home/pkappl/Programs/ADGA",
                          "dga.conf", "two_particle_4_worms.hdf5")
 x_generator = abinitio_dga.get_worm_sample_generator()
@@ -12,10 +13,5 @@ n = abinitio_dga.n_worm_samples
 f = abinitio_dga.__call__
 
 # %%
-# Do a Jackknife estimation with the x-values and f() and print the
-# results
-output = jackknife.do_jackknife_estimation(x_generator, n, f)
-print("{}{}".format("f(mean(x))             = ", output[3]))
-print("{}{}".format("mean(y')               = ", output[0]))
-print("{}{}".format("variance(y')           = ", output[1]))
-print("{}{}".format("standard deviation(y') = ", output[2]))
+# Do the Jackknife estimation
+jackknife.do_jackknife_estimation(x_generator, n, f)