Fixed compare_simplified.

It was bugged since the introduction of functional_unit

RELEASE_NOTES.md

@@ -1,7 +1,11 @@
+# 
+* Fix bug in compare_simplified introduced since the release of *functional_unit* 
+
+
 # 1.1.2
 
 Bugfix release :
-* Fix major bug #58 with axis being in diffrent order when used with multiple methods
+* Fix major bug #58 with axis being in different order when used with multiple methods
 * Fix bug #54 : unexpected behaviour with '*' wildcard and 'old_amount' parameter
 
 # 1.1

doc/source/notebooks/init.py

@@ -1,16 +1,17 @@
-import pandas as pd
 import time
-import matplotlib.pyplot as plt
-import numpy as np
+
 import brightway2 as bw
 import bw2data
 import bw2io
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
 import SALib
-from tabulate import tabulate
-from sympy import *
-from scipy.stats import binned_statistic
 import seaborn as sns
 from IPython.display import HTML, display
+from scipy.stats import binned_statistic
+from sympy import *
+from tabulate import tabulate
 
 # +
 # Custom utils defined for inter-acv
@@ -21,7 +22,6 @@
 # from lca_algebraic.stats import _generate_random_params, _compute_stochastics
 # -
 
-from IPython.display import HTML, display
 
 # Larger space in notebook for large graphs
 display(HTML("<style>.container { width:70% !important; }</style>"))

lca_algebraic/stats.py

@@ -309,7 +309,10 @@ def _stochastics(
     return problem, params, Y
 
 
-def _compute_stochastics(modelOrLambdas, methods, functional_unit=1, params=dict()):
+def _compute_stochastics(modelOrLambdas, methods, functional_unit=1, params=None):
+    if params is None:
+        params = {}
+
     if isinstance(modelOrLambdas, Activity):
         Y = compute_impacts(modelOrLambdas, methods, functional_unit=functional_unit, **params)
     else:
@@ -1292,11 +1295,11 @@ def compare_simplified(
         params, _ = _generate_random_params(100000, sample_method=StochasticMethod.RAND)
 
         # Run  Monte Carlo on full model
-        Y1 = _compute_stochastics([lambd], [method], params)
+        Y1 = _compute_stochastics([lambd], [method], params=params)
         d1 = Y1[Y1.columns[0]]
 
         # Run monte carlo of simplified model
-        Y2 = _compute_stochastics([simpl_lambd], [method], params)
+        Y2 = _compute_stochastics([simpl_lambd], [method], params=params)
         d2 = Y2[Y2.columns[0]]
 
         r_value = r_squared(Y1, Y2)