def test_write_only_activity_parameters_no_activate_others(no_init):
Database("PCB").register()
obj = ExcelImporter()
obj.db_name = "PCB"
obj.data = deepcopy(DATA)
obj.write_database(activate_parameters=False)
NEW = [{
'code':
'45cb34db4147e510a2561cceec541f6b',
'database':
'PCB',
'exchanges': [],
'name':
'unmounted printed circuit board',
'type':
'process',
'unit':
'square meter',
'parameters': [{
'name': 'something_test',
'amount': 2,
'formula': '3 + 2'
}],
}]
obj = ExcelImporter()
obj.db_name = "PCB"
obj.write_activity_parameters(NEW)
assert ActivityParameter.select().count() == 1
assert ActivityParameter.get().formula == '3 + 2'
assert 'parameters' not in get_activity(
('PCB', '45cb34db4147e510a2561cceec541f6b'))
assert 'parameters' in get_activity(
('PCB', '32aa5ab78beda5b8c8efbc89587de7a5'))
def test_compare_activities_by_grouped_leaves_html(cabgl):
result = compare_activities_by_grouped_leaves(
[bd.get_activity(("c", "6")), bd.get_activity(("c", "7"))],
("method",),
output_format="html",
)
assert isinstance(result, str)
def test_compare_activities_by_grouped_leaves_pandas(cabgl):
result = compare_activities_by_grouped_leaves(
[bd.get_activity(("c", "6")), bd.get_activity(("c", "7"))],
("method",),
output_format="pandas",
)
assert isinstance(result, pd.DataFrame)
def test_find_differences_in_inputs(fdii):
expected = {
bd.get_activity(("c", "3")): {"flow": 1},
bd.get_activity(("c", "4")): {"flow": 1.1},
bd.get_activity(("c", "5")): {"flow": 0.95},
}
assert find_differences_in_inputs(bd.get_activity(("c", "3"))) == expected
def test_compare_activities_by_grouped_leaves(cabgl):
labels, result = compare_activities_by_grouped_leaves(
[bd.get_activity(("c", "6")), bd.get_activity(("c", "7"))], ("method",)
)
assert labels == [
"activity",
"product",
"location",
"unit",
"total",
"direct emissions",
"product A",
"product B",
"product C",
"product D",
]
assert result[0][0] == "6"
assert result[1][0] == "7"
assert np.allclose(
result[0][4:],
[251, 6 / 251, (27 + 35 + 56) / 251, 56 / 251, 36 / 251, 35 / 251],
)
assert np.allclose(
result[1][4:], [212, 0, (30 + 48 + 24) / 212, 48 / 212, 32 / 212, 30 / 212]
)
def to_dataframe(self, cutoff=200):
"""Return all nonzero elements of characterized inventory as Pandas dataframe"""
assert mapping, "This method doesn't work with independent LCAs"
assert pandas, "This method requires the `pandas` (http://pandas.pydata.org/) library"
assert hasattr(
self, "characterized_inventory"), "Must do LCIA calculation first"
from bw2data import get_activity
coo = self.characterized_inventory.tocoo()
stacked = np.vstack([np.abs(coo.data), coo.row, coo.col, coo.data])
stacked.sort()
rev_activity, _, rev_bio = self.reverse_dict()
length = stacked.shape[1]
data = []
for x in range(min(cutoff, length)):
if stacked[3, length - x - 1] == 0.:
continue
activity = get_activity(rev_activity[stacked[2, length - x - 1]])
flow = get_activity(rev_bio[stacked[1, length - x - 1]])
data.append((activity['name'], flow['name'],
activity.get('location'), stacked[3, length - x - 1]))
return pandas.DataFrame(
data, columns=['Activity', 'Flow', 'Region', 'Amount'])
def test_write_only_activity_parameters_no_activate_others(no_init):
Database("PCB").register()
obj = ExcelImporter()
obj.db_name = "PCB"
obj.data = deepcopy(DATA)
obj.write_database(activate_parameters=False)
NEW = [
{
"code": "45cb34db4147e510a2561cceec541f6b",
"database": "PCB",
"exchanges": [],
"name": "unmounted printed circuit board",
"type": "process",
"unit": "square meter",
"parameters": [{"name": "something_test", "amount": 2, "formula": "3 + 2"}],
}
]
obj = ExcelImporter()
obj.db_name = "PCB"
obj.write_activity_parameters(NEW)
assert ActivityParameter.select().count() == 1
assert ActivityParameter.get().formula == "3 + 2"
assert "parameters" not in get_activity(("PCB", "45cb34db4147e510a2561cceec541f6b"))
assert "parameters" in get_activity(("PCB", "32aa5ab78beda5b8c8efbc89587de7a5"))
def test_locations_changed_as_expected(basic):
rower = Rower("animals")
rower.define_RoWs()
rower.label_RoWs()
assert get_activity(('animals', 'mutt pup'))['location'] == 'RoW_user_0'
assert get_activity(('animals', 'mutt'))['location'] == 'RoW_user_0'
assert get_activity(('animals', 'moggy'))['location'] == 'RoW_user_1'
def test_find_differences_in_inputs_locations(fdii):
expected = {
bd.get_activity(("c", "3")): {"flow": 1},
bd.get_activity(("c", "4")): {"flow": 1.1},
}
assert (
find_differences_in_inputs(bd.get_activity(("c", "3")), locations=["here"])
== expected
)
def append_one_exchange(cls, df, df_ind_j, conversion_dem_to_fu, exclude_dbs):
"""Extract info about one input activity, eg name, unit, location, etc and append it to the dataframe df."""
# Extract the activity number
k = int(''.join(c for c in df_ind_j.index[0] if c.isdigit()))
# Extract information about activity and save it
original_str = df_ind_j['DB Act {}'.format(k)]
original_db_code = df_ind_j['Activity {}'.format(k)]
# Find this input activity in brightway databases
db_name = original_db_code.split("'")[1]
code = original_db_code.split("'")[3]
original_db_code_tuple = (db_name, code)
# exclude unnecessary databases
if db_name.lower() in exclude_dbs:
return df
# Compute amount
original_on = df_ind_j['On ' + str(k)]
original_amount = df_ind_j['Amount Act ' + str(k)]
original_cfl = df_ind_j['CFL Act ' + str(k)]
computed_amount = original_on \
* original_amount \
* original_cfl \
* conversion_dem_to_fu
if db_name == 'ecoinvent 3.3 cutoff' and 'ecoinvent 3.3 cutoff' not in bd.databases:
current_project = deepcopy(bd.projects.current)
bd.projects.set_current('ecoinvent 3.3 cutoff') # Temporarily switch to ecoinvent 3.3 project
act_bw = bd.get_activity(original_db_code_tuple)
bd.projects.set_current(current_project)
input_act_values_dict = cls.create_input_act_dict(act_bw, computed_amount)
else:
try:
# Find activity using bw functionality
act_bw = bd.get_activity(original_db_code_tuple)
input_act_values_dict = cls.create_input_act_dict(act_bw, computed_amount)
except:
# If bd.get_activity does not work for whichever reason, fill info manually
# if replace_agribalyse_with_ei and "agribalyse" in db_name.lower():
# db_name = CONSUMPTION_DB_NAME TODO might need to uncomment, or fix this somewhere
input_act_values_dict = cls.bw_get_activity_info_manually(original_str, db_name, computed_amount)
# Add exchange to the dataframe with database in brightway format
input_act_values_dict.update(
{
'original_on': original_on,
'original_activity': original_db_code,
'original_db_act': original_str,
'original_cfl_act': original_cfl,
'original_amount': original_amount,
}
)
df = cls.append_exchanges_in_correct_columns(df, input_act_values_dict)
return df
def test_find_differences_in_inputs_tolerances(fdii):
assert not find_differences_in_inputs(bd.get_activity(("c", "3")), rel_tol=0.2)
expected = {
bd.get_activity(("c", "3")): {"flow": 1},
bd.get_activity(("c", "4")): {"flow": 1.1},
}
assert (
find_differences_in_inputs(bd.get_activity(("c", "3")), abs_tol=0.075)
== expected
)
def _add_results(self, res_dict, key, imp):
"""used in charachterize by flow and processes to groupby impact by year for each flow"""
if get_activity(key)['name'] not in res_dict:
#just add to dict
res_dict[get_activity(key)['name']] = imp
else:
#groupby and sum impact by year
imp[0].extend(res_dict[get_activity(key)['name']][0])
imp[1].extend(res_dict[get_activity(key)['name']][1])
c = collections.defaultdict(int)
for yr, im in zip(imp[0], imp[1]):
c[yr] += im
res_dict[get_activity(key)['name']] = tuple(
list(t) for t in zip(*sorted(zip(c.keys(), c.values())))
) #from https://stackoverflow.com/a/9764364/4929813
def test_with_ecoinvent_generic():
assert not len(Database('animals'))
animal_data = {
('animals', 'dogo'): {
'name': 'dogs',
'reference product': 'dog',
'exchanges': [],
'unit': 'kilogram',
'location': 'BR',
},
('animals', 'st bernhard'): {
'name': 'dogs',
'reference product': 'dog',
'exchanges': [],
'unit': 'kilogram',
'location': 'CH',
},
('animals', 'mutt'): {
'name': 'dogs',
'reference product': 'dog',
'exchanges': [],
'unit': 'kilogram',
'location': 'RoW',
},
}
db = Database('animals')
db.write(animal_data)
rwr = rower.Rower('animals')
rwr.load_existing(rwr.EI_GENERIC)
rwr.define_RoWs()
rwr.label_RoWs()
assert get_activity(('animals', 'mutt'))['location'] == "RoW_88"
def _update_locations_sqlite(self, mapping):
count = 0
for k, v in pyprind.prog_bar(mapping.items()):
activity = bw2data.get_activity((self.db.name, k))
activity['location'] = v
activity.save()
count += 1
return count
def compute_exiobase_weights(self):
reverse_dict = self.lca.reverse_dict()[0]
act_inds = np.where(self.d_exiobase_adjusted != 0)[0]
weights = {}
for act in act_inds:
bw_act = bd.get_activity(reverse_dict[act]).as_dict()['key']
weights[bw_act] = self.d_exiobase_adjusted[act]
return weights
def compare_to_previous(self):
if not hasattr(self, "previous_reference"):
raise ValueError("No previous reference method found")
names_found_in_data = {
get_activity(cf["input"])["name"].lower()
for ds in self.data for cf in ds["exchanges"] if cf.get("input")
}
names_missing_in_data = {
cf["name"].lower()
for ds in self.data for cf in ds["exchanges"]
if not cf.get("input")
}
names_in_reference = {
get_activity(key)["name"].lower()
for key, _ in Method(self.previous_reference).load()
}
return {
"found": names_found_in_data,
"missing": names_missing_in_data,
"reference": names_in_reference,
}
def test_compare_activities_by_grouped_leaves_max_level(cabgl):
labels, result = compare_activities_by_grouped_leaves(
[bd.get_activity(("c", "6")), bd.get_activity(("c", "7"))],
("method",),
max_level=1,
)
assert labels == [
"activity",
"product",
"location",
"unit",
"total",
"direct emissions",
"product D",
"product C",
]
assert result[0][0] == "6"
assert result[1][0] == "7"
assert np.allclose(
result[0][4:], [251, 6 / 251, (35 + 35 + 56 + 56) / 251, (36 + 27) / 251]
)
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.
"""
ra, rp, rb = lca.reverse_dict()
results = [(score, lca.inventory[index, :].sum(), rb[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 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.
"""
ra, rp, rb = lca.reverse_dict()
results = [(score, lca.supply_array[int(index)], ra[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 fix_acrolein():
ERROR = "ReCiPe 2016 only tested for ecoinvent biosphere flows; install base ecoinvent data"
assert "biosphere3" in bd.databases, ERROR
bio = bd.Database("biosphere3")
if bio.metadata.get("acrolein fixed"):
return
flow = bd.get_activity(
("biosphere3", "fa8bd05b-015d-5a82-878c-bde991551695"))
flow["CAS number"] = "107-02-8"
flow.save()
bio.metadata["acrolein fixed"] = True
bd.databases.flush()
def test_compare_activities_by_grouped_leaves_max_cutoff(cabgl):
labels, result = compare_activities_by_grouped_leaves(
[bd.get_activity(("c", "6")), bd.get_activity(("c", "7"))],
("method",),
cutoff=0.2,
)
assert labels == [
"activity",
"product",
"location",
"unit",
"total",
"direct emissions",
"product B",
"product A",
"product C",
"product D",
]
assert result[0][0] == "6"
assert result[1][0] == "7"
assert np.allclose(
result[0][4:],
[251, 6 / 251, (56 + 27 + 35) / 251, (56) / 251, 36 / 251, 35 / 251],
)
def calculate(self):
"""Calculate LCA report data"""
lca = LCA(self.activity, self.method)
lca.lci()
lca.lcia()
gt = GraphTraversal().calculate(self.activity, method=self.method)
print("FD")
force_directed = self.get_force_directed(gt["nodes"], gt["edges"], lca)
print("CA")
ca = ContributionAnalysis()
print("hinton")
hinton = ca.hinton_matrix(lca)
print("treemap")
treemap = self.get_treemap(gt["nodes"], gt["edges"], lca)
print("herfindahl")
herfindahl = herfindahl_index(lca.characterized_inventory.data)
print("concentration")
concentration = concentration_ratio(lca.characterized_inventory.data)
print("MC:")
monte_carlo = self.get_monte_carlo()
activity_data = []
for k, v in self.activity.items():
obj = get_activity(k)
activity_data.append((obj["name"], "%.2g" % v, obj["unit"]))
self.report = {
"activity": activity_data,
"method": {
"name": ": ".join(self.method),
"unit": methods[self.method]["unit"],
},
"score": float(lca.score),
"contribution": {
"hinton": hinton,
"treemap": treemap,
"herfindahl": herfindahl,
"concentration": concentration,
},
"force_directed": force_directed,
"monte carlo": monte_carlo,
"metadata": {
"type": "Brightway2 serialized LCA report",
"version": self.version,
"uuid": self.uuid,
},
}
def patch_lcia_methods(self, new_biosphere):
flows = ["PFC (CO2-eq)", "HFC (CO2-eq)"]
for flow in flows:
assert get_activity((new_biosphere, flow))
new_data = [((new_biosphere, flow), 1) for flow in flows]
count = 0
for name, metadata in methods.items():
if metadata.get("unit") == "kg CO2-Eq":
count += 1
obj = Method(name)
data = obj.load()
data.extend(new_data)
obj.write(data)
print(f"Patched {count} LCIA methods with unit 'kg CO2-Eq'")
def test_with_ecoinvent_specific_shortcut():
assert not len(Database('animals'))
animal_data = {
('animals', "6ccf7e69afcf1b74de5b52ae28bbc1c2"): {
'name': 'dogs',
'reference product': 'dog',
'exchanges': [],
'unit': 'kilogram',
'location': 'RoW',
},
}
db = Database('animals')
db.write(animal_data)
rwr = rower.Rower('animals')
rwr.apply_existing_activity_map(rwr.EI_3_4_CONSEQUENTIAL)
assert get_activity(
('animals',
"6ccf7e69afcf1b74de5b52ae28bbc1c2"))['location'] == "RoW_64"
def multi_recurse_tagged_database(activity, amount, methods, method_dicts, lca, label, default_tag, secondary_tags=[]):
if isinstance(activity, tuple):
activity = get_activity(activity)
inputs = list(activity.technosphere())
inside = [exc for exc in inputs
if exc['input'][0] == activity['database']]
outside = {exc['input']: exc['amount'] * amount
for exc in inputs
if exc['input'][0] != activity['database']}
if outside:
outside_scores = []
for n, m in enumerate(methods):
lca.switch_method(m)
lca.redo_lcia(outside)
outside_scores.append(lca.score)
else:
outside_scores = [0] * len(methods)
return {
'activity': activity,
'amount': amount,
'tag': activity.get(label) or default_tag,
'secondary_tags':[activity.get(t[0]) or t[1] for t in secondary_tags],
'impact': outside_scores,
'biosphere': [{
'amount': exc['amount'] * amount,
'impact': [exc['amount'] * amount * method_dict.get(exc['input'], 0) for method_dict in method_dicts],
'tag': exc.get(label) or activity.get(label) or default_tag,
'secondary_tags':[exc.get(t[0]) or activity.get(t[0]) or t[1] for t in secondary_tags]
} for exc in activity.biosphere()],
'technosphere': [multi_recurse_tagged_database(exc.input, exc['amount'] * amount, methods,
method_dicts, lca, label, default_tag, secondary_tags)
for exc in inside]
}
def multi_recurse_tagged_database(activity,
amount,
methods,
method_dicts,
lca,
label,
default_tag,
secondary_tags=[]):
"""Traverse a foreground database and assess activities and biosphere flows by tags using multiple methods.
Input arguments:
* ``activity``: Activity tuple or object
* ``amount``: float
* ``methods``: list of LCA methods (tuples)
* ``method_dicts``: list of dictionaries of biosphere flow tuples to CFs, e.g. ``{("biosphere", "foo"): 3}`` corresponding to methods in ``methods``
* ``lca``: An ``LCA`` object that is already initialized, i.e. has already calculated LCI
* ``label``: string
* ``default_tag``: string
* ``secondary_tags``: list of tuples in the format (secondary_label, secondary_default_tag). Default is empty list.
Returns:
.. code-block:: python
{
'activity': activity object,
'amount': float,
'tag': string,
'secondary_tags': [list of strings],
'impact': [list of floats (impact of inputs from outside foreground database) with one element per method],
'biosphere': [{
'amount': float,
'impact': [list of floats with one element per method],
'tag': string,
'secondary_tags': [list of strings]
}],
'technosphere': [this data structure]
}
"""
if isinstance(activity, tuple):
activity = get_activity(activity)
inputs = list(activity.technosphere())
inside = [exc for exc in inputs if exc["input"][0] == activity["database"]]
outside = {
exc["input"]: exc["amount"] * amount
for exc in inputs if exc["input"][0] != activity["database"]
}
if outside:
outside_scores = []
for n, m in enumerate(methods):
lca.switch_method(m)
lca.redo_lcia(outside)
outside_scores.append(lca.score)
else:
outside_scores = [0] * len(methods)
return {
"activity":
activity,
"amount":
amount,
"tag":
activity.get(label) or default_tag,
"secondary_tags": [activity.get(t[0]) or t[1] for t in secondary_tags],
"impact":
outside_scores,
"biosphere": [{
"amount":
exc["amount"] * amount,
"impact": [
exc["amount"] * amount * method_dict.get(exc["input"], 0)
for method_dict in method_dicts
],
"tag":
exc.get(label) or activity.get(label) or default_tag,
"secondary_tags": [
exc.get(t[0]) or activity.get(t[0]) or t[1]
for t in secondary_tags
],
} for exc in activity.biosphere()],
"technosphere": [
multi_recurse_tagged_database(
exc.input,
exc["amount"] * amount,
methods,
method_dicts,
lca,
label,
default_tag,
secondary_tags,
) for exc in inside
],
}
def recurse_tagged_database(activity,
amount,
method_dict,
lca,
label,
default_tag,
secondary_tags=[]):
"""Traverse a foreground database and assess activities and biosphere flows by tags.
Input arguments:
* ``activity``: Activity tuple or object
* ``amount``: float
* ``method_dict``: Dictionary of biosphere flow tuples to CFs, e.g. ``{("biosphere", "foo"): 3}``
* ``lca``: An ``LCA`` object that is already initialized, i.e. has already calculated LCI and LCIA with same method as in ``method_dict``
* ``label``: string
* ``default_tag``: string
* ``secondary_tags``: List of tuples in the format (secondary_label, secondary_default_tag). Default is empty list.
Returns:
.. code-block:: python
{
'activity': activity object,
'amount': float,
'tag': string,
'secondary_tags': [list of strings],
'impact': float (impact of inputs from outside foreground database),
'biosphere': [{
'amount': float,
'impact': float,
'tag': string,
'secondary_tags': [list of strings]
}],
'technosphere': [this data structure]
}
"""
if isinstance(activity, tuple):
activity = get_activity(activity)
inputs = list(activity.technosphere())
production = list(activity.production())
if len(production) == 1:
scale = production[0]["amount"]
elif not production:
# Assume production amount of 1
scale = 1
else:
raise ValueError("Can't scale by production exchange")
inside = [exc for exc in inputs if exc["input"][0] == activity["database"]]
outside = {
exc["input"]: exc["amount"] / scale * amount
for exc in inputs if exc["input"][0] != activity["database"]
}
if outside:
lca.redo_lcia(outside)
outside_score = lca.score
else:
outside_score = 0
return {
"activity":
activity,
"amount":
amount,
"tag":
activity.get(label) or default_tag,
"secondary_tags": [activity.get(t[0]) or t[1] for t in secondary_tags],
"impact":
outside_score,
"biosphere": [{
"amount":
exc["amount"] / scale * amount,
"impact":
exc["amount"] / scale * amount * method_dict.get(exc["input"], 0),
"tag":
exc.get(label) or activity.get(label) or default_tag,
"secondary_tags": [
exc.get(t[0]) or activity.get(t[0]) or t[1]
for t in secondary_tags
],
} for exc in activity.biosphere()],
"technosphere": [
recurse_tagged_database(
exc.input,
exc["amount"] / scale * amount,
method_dict,
lca,
label,
default_tag,
secondary_tags,
) for exc in inside
],
}
def get_name(self, key):
return get_activity(key).get("name", "Unknown")
def test_calculate_matrix_presamples():
data = {
("test-db", 'b'): {
'exchanges': [],
'type': 'emission',
},
("test-db", 't1'): {
'exchanges': [{
'amount': 1,
'input': ('test-db', 't2'),
'type': 'technosphere',
'formula': 'foo + bar'
}, {
'amount': 1,
'input': ('test-db', 'b'),
'type': 'biosphere',
'formula': 'foo - bar + pppp'
}],
'type':
'process',
},
("test-db", 't2'): {
'exchanges': [],
'type': 'process',
},
}
Database("test-db").write(data)
Group.create(name="E", order=[])
data = [{
'name': 'foo',
'database': 'test-db',
'code': 't1',
'amount': 7,
'uncertainty_type': 4,
'minimum': 0,
'maximum': 14,
}, {
'name': 'bar',
'database': 'test-db',
'code': 't1',
'amount': 11,
}]
parameters.new_project_parameters([{'name': 'pppp', 'amount': 12}])
parameters.new_activity_parameters(data, 'A')
parameters.add_exchanges_to_group('A', get_activity(("test-db", 't1')))
parameters.recalculate()
pbm = ParameterizedBrightwayModel("A")
pbm.load_parameter_data()
pbm.calculate_static()
pbm.calculate_matrix_presamples()
id_, dirpath = pbm.save_presample('test-everything')
# Check for file contents
pp = PresamplesPackage(dirpath)
resources = pp.resources
assert len(resources) == 3
assert resources[0]['type'] == 'biosphere'
assert resources[0]['samples']['shape'] == [1, 1]
assert resources[1]['type'] == 'technosphere'
assert resources[1]['samples']['shape'] == [1, 1]
assert resources[2]['label'] == 'test-everything'
assert resources[2]['samples']['shape'] == [3, 1]
def test_upstream_bio(activity):
act = get_activity(("db", "c"))
assert len(list(act.upstream())) == 0
assert len(act.upstream()) == 0