"Fossies" - the Fresh Open Source Software Archive  

Source code changes of the file "python/prophet/serialize.py" between
prophet-1.0.tar.gz and prophet-1.1.tar.gz

About: Prophet is a tool for producing high quality forecasts for time series data that has multiple seasonality with linear or non-linear growth.

serialize.py  (prophet-1.0)	:	serialize.py  (prophet-1.1)
# -*- coding: utf-8 -*-		# -*- coding: utf-8 -*-
# Copyright (c) Facebook, Inc. and its affiliates.		# Copyright (c) Facebook, Inc. and its affiliates.
# This source code is licensed under the MIT license found in the		# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.		# LICENSE file in the root directory of this source tree.
from __future__ import absolute_import, division, print_function		from __future__ import absolute_import, division, print_function
from collections import OrderedDict		from collections import OrderedDict
from copy import deepcopy		from copy import deepcopy
		from io import StringIO
import json		import json
import numpy as np		import numpy as np
import pandas as pd		import pandas as pd
from prophet.forecaster import Prophet		from prophet.forecaster import Prophet
from prophet import __version__		from prophet import __version__
SIMPLE_ATTRIBUTES = [		SIMPLE_ATTRIBUTES = [
'growth', 'n_changepoints', 'specified_changepoints', 'changepoint_range',		'growth', 'n_changepoints', 'specified_changepoints', 'changepoint_range',
skipping to change at line 39		skipping to change at line 40
PD_TIMESTAMP = ['start']		PD_TIMESTAMP = ['start']
PD_TIMEDELTA = ['t_scale']		PD_TIMEDELTA = ['t_scale']
PD_DATAFRAME = ['holidays', 'history', 'train_component_cols']		PD_DATAFRAME = ['holidays', 'history', 'train_component_cols']
NP_ARRAY = ['changepoints_t']		NP_ARRAY = ['changepoints_t']
ORDEREDDICT = ['seasonalities', 'extra_regressors']		ORDEREDDICT = ['seasonalities', 'extra_regressors']
def model_to_json(model):		def model_to_dict(model):
"""Serialize a Prophet model to json string.		"""Convert a Prophet model to a dictionary suitable for JSON serialization.
Model must be fitted. Skips Stan objects that are not needed for predict.		Model must be fitted. Skips Stan objects that are not needed for predict.
Can be deserialized with model_from_json.		Can be reversed with model_from_dict.
Parameters		Parameters
----------		----------
model: Prophet model object.		model: Prophet model object.
Returns		Returns
-------		-------
json string that can be deserialized into a Prophet model.		dict that can be used to serialize a Prophet model as JSON or loaded back
		into a Prophet model.
"""		"""
if model.history is None:		if model.history is None:
raise ValueError(		raise ValueError(
"This can only be used to serialize models that have already been fi t."		"This can only be used to serialize models that have already been fi t."
)		)
model_json = {		model_dict = {
attribute: getattr(model, attribute) for attribute in SIMPLE_ATTRIBUTES		attribute: getattr(model, attribute) for attribute in SIMPLE_ATTRIBUTES
}		}
# Handle attributes of non-core types		# Handle attributes of non-core types
for attribute in PD_SERIES:		for attribute in PD_SERIES:
if getattr(model, attribute) is None:		if getattr(model, attribute) is None:
model_json[attribute] = None		model_dict[attribute] = None
else:		else:
model_json[attribute] = getattr(model, attribute).to_json(		model_dict[attribute] = getattr(model, attribute).to_json(
orient='split', date_format='iso'		orient='split', date_format='iso'
)		)
for attribute in PD_TIMESTAMP:		for attribute in PD_TIMESTAMP:
model_json[attribute] = getattr(model, attribute).timestamp()		model_dict[attribute] = getattr(model, attribute).timestamp()
for attribute in PD_TIMEDELTA:		for attribute in PD_TIMEDELTA:
model_json[attribute] = getattr(model, attribute).total_seconds()		model_dict[attribute] = getattr(model, attribute).total_seconds()
for attribute in PD_DATAFRAME:		for attribute in PD_DATAFRAME:
if getattr(model, attribute) is None:		if getattr(model, attribute) is None:
model_json[attribute] = None		model_dict[attribute] = None
else:		else:
model_json[attribute] = getattr(model, attribute).to_json(orient='ta ble', index=False)		model_dict[attribute] = getattr(model, attribute).to_json(orient='ta ble', index=False)
for attribute in NP_ARRAY:		for attribute in NP_ARRAY:
model_json[attribute] = getattr(model, attribute).tolist()		model_dict[attribute] = getattr(model, attribute).tolist()
for attribute in ORDEREDDICT:		for attribute in ORDEREDDICT:
model_json[attribute] = [		model_dict[attribute] = [
list(getattr(model, attribute).keys()),		list(getattr(model, attribute).keys()),
getattr(model, attribute),		getattr(model, attribute),
]		]
# Other attributes with special handling		# Other attributes with special handling
# fit_kwargs -> Transform any numpy types before serializing.		# fit_kwargs -> Transform any numpy types before serializing.
# They do not need to be transformed back on deserializing.		# They do not need to be transformed back on deserializing.
fit_kwargs = deepcopy(model.fit_kwargs)		fit_kwargs = deepcopy(model.fit_kwargs)
if 'init' in fit_kwargs:		if 'init' in fit_kwargs:
for k, v in fit_kwargs['init'].items():		for k, v in fit_kwargs['init'].items():
if isinstance(v, np.ndarray):		if isinstance(v, np.ndarray):
fit_kwargs['init'][k] = v.tolist()		fit_kwargs['init'][k] = v.tolist()
elif isinstance(v, np.floating):		elif isinstance(v, np.floating):
fit_kwargs['init'][k] = float(v)		fit_kwargs['init'][k] = float(v)
model_json['fit_kwargs'] = fit_kwargs		model_dict['fit_kwargs'] = fit_kwargs
# Params (Dict[str, np.ndarray])		# Params (Dict[str, np.ndarray])
model_json['params'] = {k: v.tolist() for k, v in model.params.items()}		model_dict['params'] = {k: v.tolist() for k, v in model.params.items()}
# Attributes that are skipped: stan_fit, stan_backend		# Attributes that are skipped: stan_fit, stan_backend
model_json['__prophet_version'] = __version__		model_dict['__prophet_version'] = __version__
		return model_dict
		def model_to_json(model):
		"""Serialize a Prophet model to json string.
		Model must be fitted. Skips Stan objects that are not needed for predict.
		Can be deserialized with model_from_json.
		Parameters
		----------
		model: Prophet model object.
		Returns
		-------
		json string that can be deserialized into a Prophet model.
		"""
		model_json = model_to_dict(model)
return json.dumps(model_json)		return json.dumps(model_json)
def model_from_json(model_json):		def model_from_dict(model_dict):
"""Deserialize a Prophet model from json string.		"""Recreate a Prophet model from a dictionary.
Deserializes models that were serialized with model_to_json.		Recreates models that were converted with model_to_dict.
Parameters		Parameters
----------		----------
model_json: Serialized model string		model_dict: Dictionary containing model, created with model_to_dict.
Returns		Returns
-------		-------
Prophet model.		Prophet model.
"""		"""
attr_dict = json.loads(model_json)
model = Prophet() # We will overwrite all attributes set in init anyway		model = Prophet() # We will overwrite all attributes set in init anyway
# Simple types		# Simple types
for attribute in SIMPLE_ATTRIBUTES:		for attribute in SIMPLE_ATTRIBUTES:
setattr(model, attribute, attr_dict[attribute])		setattr(model, attribute, model_dict[attribute])
for attribute in PD_SERIES:		for attribute in PD_SERIES:
if attr_dict[attribute] is None:		if model_dict[attribute] is None:
setattr(model, attribute, None)		setattr(model, attribute, None)
else:		else:
s = pd.read_json(attr_dict[attribute], typ='series', orient='split')		s = pd.read_json(StringIO(model_dict[attribute]), typ='series', orie nt='split')
if s.name == 'ds':		if s.name == 'ds':
if len(s) == 0:		if len(s) == 0:
s = pd.to_datetime(s)		s = pd.to_datetime(s)
s = s.dt.tz_localize(None)		s = s.dt.tz_localize(None)
setattr(model, attribute, s)		setattr(model, attribute, s)
for attribute in PD_TIMESTAMP:		for attribute in PD_TIMESTAMP:
setattr(model, attribute, pd.Timestamp.utcfromtimestamp(attr_dict[attrib ute]))		setattr(model, attribute, pd.Timestamp.utcfromtimestamp(model_dict[attri bute]))
for attribute in PD_TIMEDELTA:		for attribute in PD_TIMEDELTA:
setattr(model, attribute, pd.Timedelta(seconds=attr_dict[attribute]))		setattr(model, attribute, pd.Timedelta(seconds=model_dict[attribute]))
for attribute in PD_DATAFRAME:		for attribute in PD_DATAFRAME:
if attr_dict[attribute] is None:		if model_dict[attribute] is None:
setattr(model, attribute, None)		setattr(model, attribute, None)
else:		else:
df = pd.read_json(attr_dict[attribute], typ='frame', orient='table', convert_dates=['ds'])		df = pd.read_json(StringIO(model_dict[attribute]), typ='frame', orie nt='table', convert_dates=['ds'])
if attribute == 'train_component_cols':		if attribute == 'train_component_cols':
# Special handling because of named index column		# Special handling because of named index column
df.columns.name = 'component'		df.columns.name = 'component'
df.index.name = 'col'		df.index.name = 'col'
setattr(model, attribute, df)		setattr(model, attribute, df)
for attribute in NP_ARRAY:		for attribute in NP_ARRAY:
setattr(model, attribute, np.array(attr_dict[attribute]))		setattr(model, attribute, np.array(model_dict[attribute]))
for attribute in ORDEREDDICT:		for attribute in ORDEREDDICT:
key_list, unordered_dict = attr_dict[attribute]		key_list, unordered_dict = model_dict[attribute]
od = OrderedDict()		od = OrderedDict()
for key in key_list:		for key in key_list:
od[key] = unordered_dict[key]		od[key] = unordered_dict[key]
setattr(model, attribute, od)		setattr(model, attribute, od)
# Other attributes with special handling		# Other attributes with special handling
# fit_kwargs		# fit_kwargs
model.fit_kwargs = attr_dict['fit_kwargs']		model.fit_kwargs = model_dict['fit_kwargs']
# Params (Dict[str, np.ndarray])		# Params (Dict[str, np.ndarray])
model.params = {k: np.array(v) for k, v in attr_dict['params'].items()}		model.params = {k: np.array(v) for k, v in model_dict['params'].items()}
# Skipped attributes		# Skipped attributes
model.stan_backend = None		model.stan_backend = None
model.stan_fit = None		model.stan_fit = None
return model		return model
		def model_from_json(model_json):
		"""Deserialize a Prophet model from json string.
		Deserializes models that were serialized with model_to_json.
		Parameters
		----------
		model_json: Serialized model string
		Returns
		-------
		Prophet model.
		"""
		model_dict = json.loads(model_json)
		return model_from_dict(model_dict)
End of changes. 32 change blocks.
32 lines changed or deleted		51 lines changed or added

---