sciapy.regress.models_theano.ProxyModel._lt_corr() - Code Metrics - Inspection of "ci: Install only "tests" extra for testing" - st-bender/sciapy - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Push — master ( 33eda6...e9fe96 )

by Stefan

created 2022-04-06 09:50 UTC

sciapy.regress.models_theano.ProxyModel._lt_corr() A

↳ Parent: sciapy.regress.models_theano

Complexity

Conditions

Size

Total Lines	16
Code Lines	10

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	2
eloc	10
nop	3
dl	0
loc	16
rs	9.9
c	0
b	0
f	0

# -*- coding: utf-8 -*-
# vim:fileencoding=utf-8
#
# Copyright (c) 2022 Stefan Bender
#
# This module is part of sciapy.
# sciapy is free software: you can redistribute it or modify
# it under the terms of the GNU General Public License as published
# by the Free Software Foundation, version 2.
# See accompanying LICENSE file or http://www.gnu.org/licenses/gpl-2.0.html.
"""SCIAMACHY regression models (theano/pymc3 version)

Model classes for SCIAMACHY data regression fits using
:mod:`theano` for :mod:`pymc3`.

This interface is still experimental.
"""
from __future__ import absolute_import, division, print_function
from warnings import warn

import numpy as np

try:
	import aesara_theano_fallback.tensor as tt
except ImportError as err:
	raise ImportError(
		"The `aesara_theano_fallback` package is required for the `theano` model interface."
	).with_traceback(err.__traceback__)
try:
	import pymc3 as pm
except ImportError as err:
	raise ImportError(
		"The `pymc3` package is required for the `theano` model interface."
	).with_traceback(err.__traceback__)

__all__ = [
	"HarmonicModelCosineSine", "HarmonicModelAmpPhase",
	"LifetimeModel",
	"ProxyModel",
	"ModelSet",
	"setup_proxy_model_theano",
	"trace_gas_modelset",
]


class HarmonicModelCosineSine:
	"""Model for harmonic terms

	Models harmonic terms using a cosine and sine part.
	The total amplitude and phase can be inferred from that.

	Parameters
	----------
	freq : float
		The frequency in years^-1
	cos : float
		The amplitude of the cosine part
	sin : float
		The amplitude of the sine part
	"""
	def __init__(self, freq, cos, sin):
		self.omega = tt.as_tensor_variable(2 * np.pi * freq).astype("float64")
		self.cos = tt.as_tensor_variable(cos).astype("float64")
		self.sin = tt.as_tensor_variable(sin).astype("float64")

	def get_value(self, t):
		t = tt.as_tensor_variable(t).astype("float64")
		return (
			self.cos * tt.cos(self.omega * t)
			+ self.sin * tt.sin(self.omega * t)
		)

	def get_amplitude(self):
		return tt.sqrt(self.cos**2 + self.sin**2)

	def get_phase(self):
		return tt.arctan2(self.cos, self.sin)

	def compute_gradient(self, t):
		t = tt.as_tensor_variable(t).astype("float64")
		dcos = tt.cos(self.omega * t)
		dsin = tt.sin(self.omega * t)
		df = 2 * np.pi * t * (self.sin * dcos - self.cos * dsin)
		return (df, dcos, dsin)


class HarmonicModelAmpPhase:
	"""Model for harmonic terms

	Models harmonic terms using amplitude and phase of a sine.

	Parameters
	----------
	freq : float
		The frequency in years^-1
	amp : float
		The amplitude of the harmonic term
	phase : float
		The phase of the harmonic part
	"""
	def __init__(self, freq, amp, phase):
		self.omega = tt.as_tensor_variable(2 * np.pi * freq).astype("float64")
		self.amp = tt.as_tensor_variable(amp).astype("float64")
		self.phase = tt.as_tensor_variable(phase).astype("float64")

	def get_value(self, t):
		t = tt.as_tensor_variable(t).astype("float64")
		return self.amp * tt.sin(self.omega * t + self.phase)

	def get_amplitude(self):
		return self.amp

	def get_phase(self):
		return self.phase

	def compute_gradient(self, t):
		t = tt.as_tensor_variable(t).astype("float64")
		damp = tt.sin(self.omega * t + self.phase)
		dphi = self.amp * tt.cos(self.omega * t + self.phase)
		df = 2 * np.pi * t * dphi
		return (df, damp, dphi)


class LifetimeModel:
	"""Model for variable lifetime

	Returns the positive values of the sine/cosine.

	Parameters
	----------
	harmonics : HarmonicModelCosineSine or HarmonicModelAmpPhase or list
	"""
	def __init__(self, harmonics, lower=0.):
		if not hasattr(harmonics, "getitem"):
			harmonics = [harmonics]
		self.harmonics = harmonics
		self.lower = lower

	def get_value(self, t):
		tau_cs = tt.zeros(t.shape[:-1], dtype="float64")
		for h in self.harmonics:
			tau_cs += h.get_value(t)
		return tt.maximum(self.lower, tau_cs)


def _interp(x, xs, ys, fill_value=0.):
	idx = xs.searchsorted(x)
	out_of_bounds = tt.zeros(x.shape[:-1], dtype=bool)
	out_of_bounds |= (idx < 1) | (idx >= xs.shape[0])
	idx = tt.clip(idx, 1, xs.shape[0] - 1)
	dl = x - xs[idx - 1]
	dr = xs[idx] - x
	d = dl + dr
	wl = dr / d
	ret = tt.zeros(x.shape[:-1], dtype="float64")
	ret += wl * ys[idx - 1] + (1 - wl) * ys[idx]
	ret = tt.switch(out_of_bounds, fill_value, ret)
	return ret


class ProxyModel:
	"""Model for proxy terms

	Models proxy terms with a finite and (semi-)annually varying life time.

	Parameters
	----------
	proxy_times : (N,) array_like
		The data times of the proxy values
	proxy_vals : (N,) array_like
		The proxy values at `proxy_times`
	amp : float
		The amplitude of the proxy term
	lag : float, optional
		The lag of the proxy value in years.
	tau0 : float, optional
		The base life time of the proxy
	tau_harm : LifetimeModel, optional
		The lifetime harmonic model with a lower limit.
	tau_scan : float, optional
		The number of days to sum the previous proxy values. If it is
		negative, the value will be set to three times the maximal lifetime.
		No lifetime adjustemets are calculated when set to zero.
	days_per_time_unit : float, optional
		The number of days per time unit, used to normalize the lifetime
		units. Use 365.25 if the times are in fractional years, or 1 if
		they are in days.
		Default: 365.25
	"""
	def __init__(
		self, ptimes, pvalues, amp,
		lag=0.,
		tau0=0.,
		tau_harm=None,
		tau_scan=0,
		days_per_time_unit=365.25,
	):
		# data
		self.times = tt.as_tensor_variable(ptimes).astype("float64")
		self.values = tt.as_tensor_variable(pvalues).astype("float64")
		# parameters
		self.amp = tt.as_tensor_variable(amp).astype("float64")
		self.days_per_time_unit = tt.as_tensor_variable(days_per_time_unit).astype("float64")
		self.lag = tt.as_tensor_variable(lag / days_per_time_unit).astype("float64")
		self.tau0 = tt.as_tensor_variable(tau0).astype("float64")
		self.tau_harm = tau_harm
		self.tau_scan = tau_scan
		dt = 1.0
		bs = np.arange(dt, tau_scan + dt, dt) / days_per_time_unit
		self.bs = tt.as_tensor_variable(bs).astype("float64")
		self.dt = tt.as_tensor_variable(dt).astype("float64")
		# Makes "(m)jd" and "jyear" compatible for the lifetime
		# seasonal variation. The julian epoch (the default)
		# is slightly offset with respect to (modified) julian days.
		self.t_adj = 0.
		if self.days_per_time_unit == 1:
			# discriminate between julian days and modified julian days,
			# 1.8e6 is year 216 in julian days and year 6787 in
			# modified julian days. It should be pretty safe to judge on
			# that for most use cases.
			if self.times[0] > 1.8e6:
				# julian days
				self.t_adj = 13.
			else:
				# modified julian days
				self.t_adj = -44.25
		self.t_adj = tt.as_tensor_variable(self.t_adj).astype("float64")

	def _lt_corr(self, t, tau):
		"""Lifetime corrected values

		Corrects for a finite lifetime by summing over the last `tmax`
		days with an exponential decay given of lifetime(s) `tau`.
		"""
		yp = tt.zeros(t.shape[:-1], dtype="float64")
		tauexp = tt.exp(-self.dt / tau)
		taufac = tt.ones(tau.shape[:-1], dtype="float64")
		for b in self.bs:
			taufac *= tauexp
			yp += taufac * _interp(
				t - self.lag - b,
				self.times, self.values,
			)
		return yp * self.dt

	def get_value(self, t):
		t = tt.as_tensor_variable(t)
		proxy_val = _interp(
			t - self.lag,
			self.times, self.values,
		)
		if self.tau_scan == 0:
			# no lifetime, nothing else to do
			return self.amp * proxy_val
		tau = self.tau0
		if self.tau_harm is not None:
			tau_cs = self.tau_harm.get_value(t + self.t_adj)
			tau += tau_cs
		proxy_val += self._lt_corr(t, tau)
		return self.amp * proxy_val


class ModelSet:
	def __init__(self, models):
		self.models = models

	def get_value(self, t):
		v = tt.zeros(t.shape[:-1], dtype="float64")
		for m in self.models:
			v += m.get_value(t)
		return v


def setup_proxy_model_theano(
	model, name,
	times, values,
	max_amp=1e10, max_days=100,
	**kwargs
):
	warn(
		"This method to set up the `theano`/`pymc3` interface is experimental, "
		"and the interface will most likely change in future versions. "
		"It is recommended to use the `ProxyModel` class instead."
	)
	# extract setup from `kwargs`
	fit_lag = kwargs.get("fit_lag", False)
	lag = kwargs.get("lag", 0.)
	lifetime_scan = kwargs.get("lifetime_scan", 60)
	positive = kwargs.get("positive", False)
	time_format = kwargs.get("time_format", "jyear")

	with model:
		if positive:
			log_amp = pm.Normal("log_{0}_amp".format(name), mu=0.0, sd=np.log(max_amp))
			amp = pm.Deterministic("{0}_amp".format(name), pm.math.exp(log_amp))
		else:
			amp = pm.Normal("{0}_amp".format(name), mu=0.0, sd=max_amp)
		if fit_lag:
			log_lag = pm.Normal("log_{0}_lag".format(name), mu=-5.0, sd=np.log(max_days))
			lag = pm.Deterministic("{0}_lag".format(name), pm.math.exp(log_lag))
		if lifetime_scan > 0:
			log_tau0 = pm.Normal("log_{0}_tau0".format(name), mu=-5.0, sd=np.log(max_days))
			tau0 = pm.Deterministic("{0}_tau0".format(name), pm.math.exp(log_tau0))
			cos1 = pm.Normal("{0}_tau_cos1".format(name), mu=0.0, sd=max_amp)
			sin1 = pm.Normal("{0}_tau_sin1".format(name), mu=0.0, sd=max_amp)
			harm1 = HarmonicModelCosineSine(1., cos1, sin1)
			tau1 = LifetimeModel(harm1, lower=0)
		else:
			tau0 = 0.
			tau1 = None
		proxy = ProxyModel(
			times, values,
			amp,
			lag=lag,
			tau0=tau0,
			tau_harm=tau1,
			tau_scan=lifetime_scan,
			days_per_time_unit=1 if time_format.endswith("d") else 365.25,
		)
	return proxy


def _default_proxy_config(tfmt="jyear"):

	from .load_data import load_dailymeanLya, load_dailymeanAE
	proxy_config = {}
	# Lyman-alpha
	plyat, plyadf = load_dailymeanLya(tfmt=tfmt)
	proxy_config.update({
		"Lya": {
			"times": plyat,
			"values": plyadf["Lya"],
			"lifetime_scan": 0,
			"positive": False,
		}
	})
	# AE index
	paet, paedf = load_dailymeanAE(name="GM", tfmt=tfmt)
	proxy_config.update({
		"GM": {
			"times": paet,
			"values": paedf["GM"],
			"lifetime_scan": 60,
			"positive": True,
		}
	})
	return proxy_config


def trace_gas_modelset(constant=True, freqs=None, proxy_config=None, **kwargs):
	"""Trace gas model set

	Sets up the trace gas model for easy access. All parameters are optional,
	defaults to an offset, no harmonics, proxies are uncentered and unscaled
	Lyman-alpha and AE. AE with only positive amplitude and a seasonally
	varying lifetime.

	Parameters
	----------
	constant : bool, optional
		Whether or not to include a constant (offset) term, default is True.
	freqs : list, optional
		Frequencies of the harmonic terms in 1 / a^-1 (inverse years).
	proxy_config : dict, optional
		Proxy configuration if different from the standard setup.
	**kwargs : optional
		Additional keyword arguments, all of them are also passed on to
		the proxy setup. For now, supported are the following which are
		also passed along to the proxy setup with
		`setup_proxy_model_with_bounds()`:

		* fit_phase : bool
			fit amplitude and phase instead of sine and cosine
		* scale : float
			the factor by which the data is scaled, used to constrain
			the maximum and minimum amplitudes to be fitted.
		* time_format : string
			The `astropy.time.Time` format string to setup the time axis.
		* days_per_time_unit : float
			The number of days per time unit, used to normalize the frequencies
			for the harmonic terms. Use 365.25 if the times are in fractional years,
			1 if they are in days. Default: 365.25
		* max_amp : float
			Maximum magnitude of the coefficients, used to constrain the
			parameter search.
		* max_days : float
			Maximum magnitude of the lifetimes, used to constrain the
			parameter search.

	Returns
	-------
	model : :class:`TraceGasModelSet` (extends :class:`celerite.ModelSet`)
	"""
	warn(
		"This method to set up the `theano`/`pymc3` interface is experimental, "
		"and the interface will most likely change in future versions. "
		"It is recommended to use the `ProxyModel` class instead."
	)
	fit_phase = kwargs.get("fit_phase", False)
	scale = kwargs.get("scale", 1e-6)
	tfmt = kwargs.get("time_format", "jyear")
	delta_t = kwargs.get("days_per_time_unit", 365.25)

	max_amp = kwargs.pop("max_amp", 1e10 * scale)
	max_days = kwargs.pop("max_days", 100)

	proxy_config = proxy_config or _default_proxy_config(tfmt=tfmt)

	with pm.Model() as model:
		offset = 0.
		if constant:
			offset = pm.Normal("offset", mu=0.0, sd=max_amp)

		modelset = []
		for freq in freqs:
			if not fit_phase:
				cos = pm.Normal("cos{0}".format(freq), mu=0., sd=max_amp)
				sin = pm.Normal("sin{0}".format(freq), mu=0., sd=max_amp)
				harm = HarmonicModelCosineSine(
					freq * delta_t / 365.25,
					cos, sin,
				)
			else:
				amp = pm.Normal("amp{0}".format(freq), mu=0., sd=max_amp)
				phase = pm.Normal("phase{0}".format(freq), mu=0., sd=max_amp)
				harm = HarmonicModelAmpPhase(
					freq * delta_t / 365.25,
					amp, phase,
				)
			modelset.append(harm)

		for pn, conf in proxy_config.items():
			if "max_amp" not in conf:
				conf.update(dict(max_amp=max_amp))
			if "max_days" not in conf:
				conf.update(dict(max_days=max_days))
			kw = kwargs.copy()  # don't mess with the passed arguments
			kw.update(conf)
			modelset.append(
				setup_proxy_model_theano(model, pn, **kw)
			)

	return model, ModelSet(modelset), offset


1		# -- coding: utf-8 --
2		# vim:fileencoding=utf-8
3		#
4		# Copyright (c) 2022 Stefan Bender
5		#
6		# This module is part of sciapy.
7		# sciapy is free software: you can redistribute it or modify
8		# it under the terms of the GNU General Public License as published
9		# by the Free Software Foundation, version 2.
10		# See accompanying LICENSE file or http://www.gnu.org/licenses/gpl-2.0.html.
11		"""SCIAMACHY regression models (theano/pymc3 version)
12
13		Model classes for SCIAMACHY data regression fits using
14		:mod:`theano` for :mod:`pymc3`.
15
16		This interface is still experimental.
17		"""
18		from __future__ import absolute_import, division, print_function
19		from warnings import warn
20
21		import numpy as np
22
23		try:
24		import aesara_theano_fallback.tensor as tt
25		except ImportError as err:
26		raise ImportError(
27		"The `aesara_theano_fallback` package is required for the `theano` model interface."
28		).with_traceback(err.__traceback__)
29		try:
30		import pymc3 as pm
31		except ImportError as err:
32		raise ImportError(
33		"The `pymc3` package is required for the `theano` model interface."
34		).with_traceback(err.__traceback__)
35
36		__all__ = [
37		"HarmonicModelCosineSine", "HarmonicModelAmpPhase",
38		"LifetimeModel",
39		"ProxyModel",
40		"ModelSet",
41		"setup_proxy_model_theano",
42		"trace_gas_modelset",
43		]
44
45
46		class HarmonicModelCosineSine:
47		"""Model for harmonic terms
48
49		Models harmonic terms using a cosine and sine part.
50		The total amplitude and phase can be inferred from that.
51
52		Parameters
53		----------
54		freq : float
55		The frequency in years^-1
56		cos : float
57		The amplitude of the cosine part
58		sin : float
59		The amplitude of the sine part
60		"""
61		def __init__(self, freq, cos, sin):
62		self.omega = tt.as_tensor_variable(2 * np.pi * freq).astype("float64")
63		self.cos = tt.as_tensor_variable(cos).astype("float64")
64		self.sin = tt.as_tensor_variable(sin).astype("float64")
65
66		def get_value(self, t):
67		t = tt.as_tensor_variable(t).astype("float64")
68		return (
69		self.cos * tt.cos(self.omega * t)
70		+ self.sin * tt.sin(self.omega * t)
71		)
72
73		def get_amplitude(self):
74		return tt.sqrt(self.cos2 + self.sin2)
75
76		def get_phase(self):
77		return tt.arctan2(self.cos, self.sin)
78
79		def compute_gradient(self, t):
80		t = tt.as_tensor_variable(t).astype("float64")
81		dcos = tt.cos(self.omega * t)
82		dsin = tt.sin(self.omega * t)
83		df = 2 * np.pi * t * (self.sin * dcos - self.cos * dsin)
84		return (df, dcos, dsin)
85
86
87		class HarmonicModelAmpPhase:
88		"""Model for harmonic terms
89
90		Models harmonic terms using amplitude and phase of a sine.
91
92		Parameters
93		----------
94		freq : float
95		The frequency in years^-1
96		amp : float
97		The amplitude of the harmonic term
98		phase : float
99		The phase of the harmonic part
100		"""
101		def __init__(self, freq, amp, phase):
102		self.omega = tt.as_tensor_variable(2 * np.pi * freq).astype("float64")
103		self.amp = tt.as_tensor_variable(amp).astype("float64")
104		self.phase = tt.as_tensor_variable(phase).astype("float64")
105
106		def get_value(self, t):
107		t = tt.as_tensor_variable(t).astype("float64")
108		return self.amp * tt.sin(self.omega * t + self.phase)
109
110		def get_amplitude(self):
111		return self.amp
112
113		def get_phase(self):
114		return self.phase
115
116		def compute_gradient(self, t):
117		t = tt.as_tensor_variable(t).astype("float64")
118		damp = tt.sin(self.omega * t + self.phase)
119		dphi = self.amp * tt.cos(self.omega * t + self.phase)
120		df = 2 * np.pi * t * dphi
121		return (df, damp, dphi)
122
123
124		class LifetimeModel:
125		"""Model for variable lifetime
126
127		Returns the positive values of the sine/cosine.
128
129		Parameters
130		----------
131		harmonics : HarmonicModelCosineSine or HarmonicModelAmpPhase or list
132		"""
133		def __init__(self, harmonics, lower=0.):
134		if not hasattr(harmonics, "getitem"):
135		harmonics = [harmonics]
136		self.harmonics = harmonics
137		self.lower = lower
138
139		def get_value(self, t):
140		tau_cs = tt.zeros(t.shape[:-1], dtype="float64")
141		for h in self.harmonics:
142		tau_cs += h.get_value(t)
143		return tt.maximum(self.lower, tau_cs)
144
145
146		def _interp(x, xs, ys, fill_value=0.):
147		idx = xs.searchsorted(x)
148		out_of_bounds = tt.zeros(x.shape[:-1], dtype=bool)
149		out_of_bounds \|= (idx < 1) \| (idx >= xs.shape[0])
150		idx = tt.clip(idx, 1, xs.shape[0] - 1)
151		dl = x - xs[idx - 1]
152		dr = xs[idx] - x
153		d = dl + dr
154		wl = dr / d
155		ret = tt.zeros(x.shape[:-1], dtype="float64")
156		ret += wl * ys[idx - 1] + (1 - wl) * ys[idx]
157		ret = tt.switch(out_of_bounds, fill_value, ret)
158		return ret
159
160
161		class ProxyModel:
162		"""Model for proxy terms
163
164		Models proxy terms with a finite and (semi-)annually varying life time.
165
166		Parameters
167		----------
168		proxy_times : (N,) array_like
169		The data times of the proxy values
170		proxy_vals : (N,) array_like
171		The proxy values at `proxy_times`
172		amp : float
173		The amplitude of the proxy term
174		lag : float, optional
175		The lag of the proxy value in years.
176		tau0 : float, optional
177		The base life time of the proxy
178		tau_harm : LifetimeModel, optional
179		The lifetime harmonic model with a lower limit.
180		tau_scan : float, optional
181		The number of days to sum the previous proxy values. If it is
182		negative, the value will be set to three times the maximal lifetime.
183		No lifetime adjustemets are calculated when set to zero.
184		days_per_time_unit : float, optional
185		The number of days per time unit, used to normalize the lifetime
186		units. Use 365.25 if the times are in fractional years, or 1 if
187		they are in days.
188		Default: 365.25
189		"""
190		def __init__(
191		self, ptimes, pvalues, amp,
192		lag=0.,
193		tau0=0.,
194		tau_harm=None,
195		tau_scan=0,
196		days_per_time_unit=365.25,
197		):
198		# data
199		self.times = tt.as_tensor_variable(ptimes).astype("float64")
200		self.values = tt.as_tensor_variable(pvalues).astype("float64")
201		# parameters
202		self.amp = tt.as_tensor_variable(amp).astype("float64")
203		self.days_per_time_unit = tt.as_tensor_variable(days_per_time_unit).astype("float64")
204		self.lag = tt.as_tensor_variable(lag / days_per_time_unit).astype("float64")
205		self.tau0 = tt.as_tensor_variable(tau0).astype("float64")
206		self.tau_harm = tau_harm
207		self.tau_scan = tau_scan
208		dt = 1.0
209		bs = np.arange(dt, tau_scan + dt, dt) / days_per_time_unit
210		self.bs = tt.as_tensor_variable(bs).astype("float64")
211		self.dt = tt.as_tensor_variable(dt).astype("float64")
212		# Makes "(m)jd" and "jyear" compatible for the lifetime
213		# seasonal variation. The julian epoch (the default)
214		# is slightly offset with respect to (modified) julian days.
215		self.t_adj = 0.
216		if self.days_per_time_unit == 1:
217		# discriminate between julian days and modified julian days,
218		# 1.8e6 is year 216 in julian days and year 6787 in
219		# modified julian days. It should be pretty safe to judge on
220		# that for most use cases.
221		if self.times[0] > 1.8e6:
222		# julian days
223		self.t_adj = 13.
224		else:
225		# modified julian days
226		self.t_adj = -44.25
227		self.t_adj = tt.as_tensor_variable(self.t_adj).astype("float64")
228
229		def _lt_corr(self, t, tau):
230		"""Lifetime corrected values
231
232		Corrects for a finite lifetime by summing over the last `tmax`
233		days with an exponential decay given of lifetime(s) `tau`.
234		"""
235		yp = tt.zeros(t.shape[:-1], dtype="float64")
236		tauexp = tt.exp(-self.dt / tau)
237		taufac = tt.ones(tau.shape[:-1], dtype="float64")
238		for b in self.bs:
239		taufac *= tauexp
240		yp += taufac * _interp(
241		t - self.lag - b,
242		self.times, self.values,
243		)
244		return yp * self.dt
245
246		def get_value(self, t):
247		t = tt.as_tensor_variable(t)
248		proxy_val = _interp(
249		t - self.lag,
250		self.times, self.values,
251		)
252		if self.tau_scan == 0:
253		# no lifetime, nothing else to do
254		return self.amp * proxy_val
255		tau = self.tau0
256		if self.tau_harm is not None:
257		tau_cs = self.tau_harm.get_value(t + self.t_adj)
258		tau += tau_cs
259		proxy_val += self._lt_corr(t, tau)
260		return self.amp * proxy_val
261
262
263		class ModelSet:
264		def __init__(self, models):
265		self.models = models
266
267		def get_value(self, t):
268		v = tt.zeros(t.shape[:-1], dtype="float64")
269		for m in self.models:
270		v += m.get_value(t)
271		return v
272
273
274		def setup_proxy_model_theano(
275		model, name,
276		times, values,
277		max_amp=1e10, max_days=100,
278		**kwargs
279		):
280		warn(
281		"This method to set up the `theano`/`pymc3` interface is experimental, "
282		"and the interface will most likely change in future versions. "
283		"It is recommended to use the `ProxyModel` class instead."
284		)
285		# extract setup from `kwargs`
286		fit_lag = kwargs.get("fit_lag", False)
287		lag = kwargs.get("lag", 0.)
288		lifetime_scan = kwargs.get("lifetime_scan", 60)
289		positive = kwargs.get("positive", False)
290		time_format = kwargs.get("time_format", "jyear")
291
292		with model:
293		if positive:
294		log_amp = pm.Normal("log_{0}_amp".format(name), mu=0.0, sd=np.log(max_amp))
295		amp = pm.Deterministic("{0}_amp".format(name), pm.math.exp(log_amp))
296		else:
297		amp = pm.Normal("{0}_amp".format(name), mu=0.0, sd=max_amp)
298		if fit_lag:
299		log_lag = pm.Normal("log_{0}_lag".format(name), mu=-5.0, sd=np.log(max_days))
300		lag = pm.Deterministic("{0}_lag".format(name), pm.math.exp(log_lag))
301		if lifetime_scan > 0:
302		log_tau0 = pm.Normal("log_{0}_tau0".format(name), mu=-5.0, sd=np.log(max_days))
303		tau0 = pm.Deterministic("{0}_tau0".format(name), pm.math.exp(log_tau0))
304		cos1 = pm.Normal("{0}_tau_cos1".format(name), mu=0.0, sd=max_amp)
305		sin1 = pm.Normal("{0}_tau_sin1".format(name), mu=0.0, sd=max_amp)
306		harm1 = HarmonicModelCosineSine(1., cos1, sin1)
307		tau1 = LifetimeModel(harm1, lower=0)
308		else:
309		tau0 = 0.
310		tau1 = None
311		proxy = ProxyModel(
312		times, values,
313		amp,
314		lag=lag,
315		tau0=tau0,
316		tau_harm=tau1,
317		tau_scan=lifetime_scan,
318		days_per_time_unit=1 if time_format.endswith("d") else 365.25,
319		)
320		return proxy
321
322
323	View Code Duplication	def _default_proxy_config(tfmt="jyear"):
		0 ignored issues – show Duplication introduced 2019-02-21 16:06 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
324		from .load_data import load_dailymeanLya, load_dailymeanAE
325		proxy_config = {}
326		# Lyman-alpha
327		plyat, plyadf = load_dailymeanLya(tfmt=tfmt)
328		proxy_config.update({
329		"Lya": {
330		"times": plyat,
331		"values": plyadf["Lya"],
332		"lifetime_scan": 0,
333		"positive": False,
334		}
335		})
336		# AE index
337		paet, paedf = load_dailymeanAE(name="GM", tfmt=tfmt)
338		proxy_config.update({
339		"GM": {
340		"times": paet,
341		"values": paedf["GM"],
342		"lifetime_scan": 60,
343		"positive": True,
344		}
345		})
346		return proxy_config
347
348
349		def trace_gas_modelset(constant=True, freqs=None, proxy_config=None, **kwargs):
350		"""Trace gas model set
351
352		Sets up the trace gas model for easy access. All parameters are optional,
353		defaults to an offset, no harmonics, proxies are uncentered and unscaled
354		Lyman-alpha and AE. AE with only positive amplitude and a seasonally
355		varying lifetime.
356
357		Parameters
358		----------
359		constant : bool, optional
360		Whether or not to include a constant (offset) term, default is True.
361		freqs : list, optional
362		Frequencies of the harmonic terms in 1 / a^-1 (inverse years).
363		proxy_config : dict, optional
364		Proxy configuration if different from the standard setup.
365		**kwargs : optional
366		Additional keyword arguments, all of them are also passed on to
367		the proxy setup. For now, supported are the following which are
368		also passed along to the proxy setup with
369		`setup_proxy_model_with_bounds()`:
370
371		* fit_phase : bool
372		fit amplitude and phase instead of sine and cosine
373		* scale : float
374		the factor by which the data is scaled, used to constrain
375		the maximum and minimum amplitudes to be fitted.
376		* time_format : string
377		The `astropy.time.Time` format string to setup the time axis.
378		* days_per_time_unit : float
379		The number of days per time unit, used to normalize the frequencies
380		for the harmonic terms. Use 365.25 if the times are in fractional years,
381		1 if they are in days. Default: 365.25
382		* max_amp : float
383		Maximum magnitude of the coefficients, used to constrain the
384		parameter search.
385		* max_days : float
386		Maximum magnitude of the lifetimes, used to constrain the
387		parameter search.
388
389		Returns
390		-------
391		model : :class:`TraceGasModelSet` (extends :class:`celerite.ModelSet`)
392		"""
393		warn(
394		"This method to set up the `theano`/`pymc3` interface is experimental, "
395		"and the interface will most likely change in future versions. "
396		"It is recommended to use the `ProxyModel` class instead."
397		)
398		fit_phase = kwargs.get("fit_phase", False)
399		scale = kwargs.get("scale", 1e-6)
400		tfmt = kwargs.get("time_format", "jyear")
401		delta_t = kwargs.get("days_per_time_unit", 365.25)
402
403		max_amp = kwargs.pop("max_amp", 1e10 * scale)
404		max_days = kwargs.pop("max_days", 100)
405
406		proxy_config = proxy_config or _default_proxy_config(tfmt=tfmt)
407
408		with pm.Model() as model:
409		offset = 0.
410		if constant:
411		offset = pm.Normal("offset", mu=0.0, sd=max_amp)
412
413		modelset = []
414		for freq in freqs:
415		if not fit_phase:
416		cos = pm.Normal("cos{0}".format(freq), mu=0., sd=max_amp)
417		sin = pm.Normal("sin{0}".format(freq), mu=0., sd=max_amp)
418		harm = HarmonicModelCosineSine(
419		freq * delta_t / 365.25,
420		cos, sin,
421		)
422		else:
423		amp = pm.Normal("amp{0}".format(freq), mu=0., sd=max_amp)
424		phase = pm.Normal("phase{0}".format(freq), mu=0., sd=max_amp)
425		harm = HarmonicModelAmpPhase(
426		freq * delta_t / 365.25,
427		amp, phase,
428		)
429		modelset.append(harm)
430
431		for pn, conf in proxy_config.items():
432		if "max_amp" not in conf:
433		conf.update(dict(max_amp=max_amp))
434		if "max_days" not in conf:
435		conf.update(dict(max_days=max_days))
436		kw = kwargs.copy() # don't mess with the passed arguments
437		kw.update(conf)
438		modelset.append(
439		setup_proxy_model_theano(model, pn, **kw)
440		)
441
442		return model, ModelSet(modelset), offset
443

st-bender / sciapy

Push — master ( 33eda6...e9fe96 )

sciapy.regress.models_theano.ProxyModel._lt_corr() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like