def save_step_to_disk(step):
"""
Save a model step to disk, over-writing the previous file. The file will be named
'model-name.yaml' and will be saved to the initialization directory.
"""
if _disk_store is None:
print(
"Please run 'modelmanager.initialize()' before registering new model steps"
)
return
print("Saving '{}.yaml': {}".format(step.name,
os.path.join(os.getcwd(),
_disk_store)))
d = step.to_dict()
# Save supplemental objects
if 'supplemental_objects' in d:
for item in filter(None, d['supplemental_objects']):
save_supplemental_object(step.name, **item)
del item['content']
# Save main yaml file
headers = {'modelmanager_version': __version__}
content = OrderedDict(headers)
content.update({'saved_object': d})
yamlio.convert_to_yaml(content,
os.path.join(_disk_store, step.name + '.yaml'))
def save_steps_to_disk():
"""
Save current representation of model steps to disk, over-writing the previous file.
"""
headers = {'modelmanager_version': MODELMANAGER_VERSION}
content = OrderedDict(headers)
for k in sorted(_STEPS.keys()):
content.update({k: _STEPS[k]})
yamlio.convert_to_yaml(content, DISK_STORE)
def from_dict(cls, d):
"""
Create an object instance from a saved dictionary representation.
Parameters
----------
d : dict
Returns
-------
OLSRegressionStep
"""
# Pass values from the dictionary to the __init__() method
obj = cls(tables=d['tables'],
model_expression=d['model_expression'],
filters=d['filters'],
out_tables=d['out_tables'],
out_column=d['out_column'],
out_transform=d['out_transform'],
out_filters=d['out_filters'],
name=d['name'],
tags=d['tags'])
obj.summary_table = d['summary_table']
obj.fitted_parameters = d['fitted_parameters']
# Unpack the urbansim.models.RegressionModel() sub-object and resuscitate it
model_config = yamlio.convert_to_yaml(d['model'], None)
obj.model = RegressionModel.from_yaml(model_config)
return obj
def save_step(d):
"""
Save a model step to disk, over-writing the previous file. The file will be named
'model-name.yaml' and will be saved to the initialization directory.
Note: This function is for internal use. In a model building workflow, it's better to
use an object's `register()` method to register it with ModelManager and save it to
disk at the same time.
"""
if _DISK_STORE is None:
print("Please run 'mm.initialize()' before registering new model steps")
return
print("Saving '{}.yaml': {}".format(d['name'],
os.path.join(os.getcwd(), _DISK_STORE)))
headers = {'modelmanager_version': MODELMANAGER_VERSION}
content = OrderedDict(headers)
content.update({'saved_object': d})
yamlio.convert_to_yaml(content, os.path.join(_DISK_STORE, d['name']+'.yaml'))
def to_yaml(self, str_or_buffer=None):
"""
Save a model respresentation to YAML.
Parameters
----------
str_or_buffer : str or file like, optional
By default a YAML string is returned. If a string is
given here the YAML will be written to that file.
If an object with a ``.write`` method is given the
YAML will be written to that object.
Returns
-------
j : str
YAML is string if `str_or_buffer` is not given.
"""
logger.debug('serializing LCM model {} to YAML'.format(self.name))
#if (not isinstance(self.probability_mode, str) or
# not isinstance(self.choice_mode, str)):
# raise TypeError(
# 'Cannot serialize model with non-string probability_mode '
# 'or choice_mode attributes.')
return yamlio.convert_to_yaml(self.to_dict(), str_or_buffer)
def create_proforma_config(proforma_settings):
yaml_file = misc.config("proforma_user.yaml")
user_cfg = yamlio.yaml_to_dict(str_or_buffer=yaml_file)
config = psrcdev.update_sqftproforma(user_cfg, proforma_settings)
yamlio.convert_to_yaml(config, "proforma.yaml")