-
Notifications
You must be signed in to change notification settings - Fork 13
/
contribution.py
215 lines (177 loc) · 7.95 KB
/
contribution.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
import numpy as np
from bw2data import get_activity
class ContributionAnalysis:
def sort_array(self, data, limit=25, limit_type="number", total=None):
"""
Common sorting function for all ``top`` methods. Sorts by highest value first.
Operates in either ``number`` or ``percent`` mode. In ``number`` mode, return ``limit`` values. In ``percent`` mode, return all values >= (total * limit); where ``0 < limit <= 1``.
Returns 2-d numpy array of sorted values and row indices, e.g.:
.. code-block:: python
ContributionAnalysis().sort_array((1., 3., 2.))
returns
.. code-block:: python
(
(3, 1),
(2, 2),
(1, 0)
)
Args:
* *data* (numpy array): A 1-d array of values to sort.
* *limit* (number, default=25): Number of values to return, or percentage cutoff.
* *limit_type* (str, default=``number``): Either ``number`` or ``percent``.
* *total* (number, default=None): Optional specification of summed data total.
Returns:
2-d numpy array of values and row indices.
"""
total = total or np.abs(data).sum()
if limit_type not in ("number", "percent"):
raise ValueError("limit_type must be either 'percent' or 'index'.")
if limit_type == "percent":
if not 0 < limit <= 1:
raise ValueError("Percentage limits > 0 and <= 1.")
limit = (np.abs(data) >= (total * limit)).sum()
results = np.hstack(
(data.reshape((-1, 1)), np.arange(data.shape[0]).reshape((-1, 1)))
)
return results[np.argsort(np.abs(data))[::-1]][:limit, :]
def top_matrix(self, matrix, rows=5, cols=5):
"""
Find most important (i.e. highest summed) rows and columns in a matrix, as well as the most corresponding non-zero individual elements in the top rows and columns.
Only returns matrix values which are in the top rows and columns. Element values are returned as a tuple: ``(row, col, row index in top rows, col index in top cols, value)``.
Example:
.. code-block:: python
matrix = [
[0, 0, 1, 0],
[2, 0, 4, 0],
[3, 0, 1, 1],
[0, 7, 0, 1],
]
In this matrix, the row sums are ``(1, 6, 5, 8)``, and the columns sums are ``(5, 7, 6, 2)``. Therefore, the top rows are ``(3, 1)`` and the top columns are ``(1, 2)``. The result would therefore be:
.. code-block:: python
(
(
(3, 1, 0, 0, 7),
(3, 2, 0, 1, 1),
(1, 2, 1, 1, 4)
),
(3, 1),
(1, 2)
)
Args:
* *matrix* (array or matrix): Any Python object that supports the ``.sum(axis=)`` syntax.
* *rows* (int): Number of rows to select.
* *cols* (int): Number of columns to select.
Returns:
(elements, top rows, top columns)
"""
top_rows = np.argsort(np.abs(np.array(matrix.sum(axis=1)).ravel()))[
: -rows - 1 : -1
]
top_cols = np.argsort(np.abs(np.array(matrix.sum(axis=0)).ravel()))[
: -cols - 1 : -1
]
elements = []
for row, x in enumerate(top_rows):
for col, y in enumerate(top_cols):
if matrix[x, y] != 0:
elements.append((x, y, row, col, float(matrix[x, y])))
return elements, top_rows.astype(int), top_cols.astype(int)
def hinton_matrix(self, lca, rows=5, cols=5):
coo, b, t = self.top_matrix(lca.characterized_inventory, rows=rows, cols=cols)
coo = [row[2:] for row in coo] # Don't need matrix indices
flows = [self.get_name(lca.dicts.biosphere.reversed[x]) for x in b]
activities = [self.get_name(lca.dicts.activity.reversed[x]) for x in t]
return {
"results": coo,
"total": lca.score,
"xlabels": activities,
"ylabels": flows,
}
def annotate(self, sorted_data, rev_mapping):
"""Reverse the mapping from database ids to array indices"""
return [(row[0], rev_mapping[row[1]]) for row in sorted_data]
def top_processes(self, matrix, **kwargs):
"""Return an array of [value, index] technosphere processes."""
return self.sort_array(np.array(matrix.sum(axis=0)).ravel(), **kwargs)
def top_emissions(self, matrix, **kwargs):
"""Return an array of [value, index] biosphere emissions."""
return self.sort_array(np.array(matrix.sum(axis=1)).ravel(), **kwargs)
def annotated_top_processes(self, lca, names=True, **kwargs):
"""Get list of most damaging processes in an LCA, sorted by ``abs(direct impact)``.
Returns a list of tuples: ``(lca score, supply, activity)``. If ``names`` is False, they returns the process key as the last element.
"""
results = [
(
score,
lca.supply_array[int(index)],
lca.dicts.activity.reversed[int(index)],
)
for score, index in self.top_processes(
lca.characterized_inventory, **kwargs
)
]
if names:
results = [(x[0], x[1], get_activity(x[2])) for x in results]
return results
def annotated_top_emissions(self, lca, names=True, **kwargs):
"""Get list of most damaging biosphere flows in an LCA, sorted by ``abs(direct impact)``.
Returns a list of tuples: ``(lca score, inventory amount, activity)``. If ``names`` is False, they returns the process key as the last element.
"""
results = [
(score, lca.inventory[int(index), :].sum(), lca.dicts.biosphere.reversed[int(index)])
for score, index in self.top_emissions(
lca.characterized_inventory, **kwargs
)
]
if names:
results = [(x[0], x[1], get_activity(x[2])) for x in results]
return results
def get_name(self, key):
return get_activity(key).get("name", "Unknown")
def d3_treemap(
self, matrix, rev_bio, rev_techno, limit=0.025, limit_type="percent"
):
"""
Construct treemap input data structure for LCA result. Output like:
.. code-block:: python
{
"name": "LCA result",
"children": [{
"name": process 1,
"children": [
{"name": emission 1, "size": score},
{"name": emission 2, "size": score},
],
}]
}
"""
total = np.abs(matrix).sum()
processes = self.top_processes(matrix, limit=limit, limit_type=limit_type)
data = {"name": "LCA result", "children": [], "size": total}
for _, tech_index in processes:
name = self.get_name(rev_techno[tech_index])
this_score = np.abs(matrix[:, int(tech_index)].toarray().ravel()).sum()
children = []
for score, bio_index in self.sort_array(
matrix[:, int(tech_index)].toarray().ravel(),
limit=limit,
limit_type=limit_type,
total=total,
):
children.append(
{
"name": self.get_name(rev_bio[bio_index]),
"size": float(abs(matrix[int(bio_index), int(tech_index)])),
}
)
children_score = sum([x["size"] for x in children])
if children_score < (0.95 * this_score):
children.append({"name": "Others", "size": this_score - children_score})
data["children"].append(
{
"name": name,
"size": this_score,
# "children": children
}
)
return data