pyelq 1.1.3__tar.gz → 1.2.0__tar.gz

{pyelq-1.1.3 → pyelq-1.2.0}/PKG-INFO

@@ -1,27 +1,28 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: pyelq
-Version: 1.1.3
+Version: 1.2.0
 Summary: Package for detection, localization and quantification code.
 License: Apache-2.0
+License-File: LICENSE.md
+License-File: LICENSES/Apache-2.0.txt
 Keywords: gas dispersion,emission,detection,localization,quantification
 Author: Bas van de Kerkhof
-Requires-Python: >=3.9
+Requires-Python: >=3.11
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Requires-Dist: geojson (>=3.2.0)
-Requires-Dist: numpy (>=2.0.2)
-Requires-Dist: openmcmc (==1.0.6)
-Requires-Dist: pandas (>=2.2.3)
-Requires-Dist: plotly (>=6.0.0)
-Requires-Dist: pymap3d (>=3.1.0)
-Requires-Dist: scikit-learn (>=1.6.1)
-Requires-Dist: scipy (>=1.13.1)
-Requires-Dist: shapely (>=2.0.7)
+Requires-Dist: numpy (>=2.4.0)
+Requires-Dist: openmcmc (==1.0.8)
+Requires-Dist: pandas (>=2.3.3,<3.0.0)
+Requires-Dist: plotly (>=6.5.0)
+Requires-Dist: pymap3d (>=3.2.0)
+Requires-Dist: scikit-learn (>=1.8.0)
+Requires-Dist: scipy (>=1.16.3)
+Requires-Dist: shapely (>=2.1.2)
 Project-URL: Documentation, https://sede-open.github.io/pyELQ/
 Project-URL: Homepage, https://sede-open.github.io/pyELQ/
 Project-URL: Repository, https://github.com/sede-open/pyELQ
@@ -51,27 +52,27 @@ SPDX-License-Identifier: Apache-2.0
 </div>
 
 # pyELQ
-This repository contains the Python Emission Localization and Quantification software we call pyELQ. It is code used 
-for gas dispersion modelling, in particular methane emissions detection, localization and quantification. 
+This repository contains the Python Emission Localization and Quantification software we call pyELQ. It is code used
+for gas dispersion modelling, in particular methane emissions detection, localization and quantification.
 
 ***
 # Background
-The **py**thon **E**mission **L**ocalization and **Q**uantification (pyELQ) code aims to maximize effective use of 
-existing measurement data, especially from continuous monitoring solutions. The code has been developed to detect, 
-localize, and quantify methane emissions from concentration and wind measurements. It can be used in combination with 
+The **py**thon **E**mission **L**ocalization and **Q**uantification (pyELQ) code aims to maximize effective use of
+existing measurement data, especially from continuous monitoring solutions. The code has been developed to detect,
+localize, and quantify methane emissions from concentration and wind measurements. It can be used in combination with
 point or beam sensors that are placed strategically on an area of interest.
 
-The algorithms in the pyELQ code are based a Bayesian statistics framework. pyELQ can ingest long-term concentration 
-and wind data, and it performs an inversion to predict the likely strengths and locations of persistent methane sources. 
-The goal is to arrive at a plausible estimate of methane emissions from an area of interest that matches the measured 
-data. The predictions from pyELQ come with uncertainty ranges that are representative of probability density functions 
-sampled by a Markov Chain Monte Carlo method. Time series of varying length can be processed by pyELQ: in general, 
-the Bayesian inversion leads to a more constrained solution if more high-precision measurement data is available. 
+The algorithms in the pyELQ code are based a Bayesian statistics framework. pyELQ can ingest long-term concentration
+and wind data, and it performs an inversion to predict the likely strengths and locations of persistent methane sources.
+The goal is to arrive at a plausible estimate of methane emissions from an area of interest that matches the measured
+data. The predictions from pyELQ come with uncertainty ranges that are representative of probability density functions
+sampled by a Markov Chain Monte Carlo method. Time series of varying length can be processed by pyELQ: in general,
+the Bayesian inversion leads to a more constrained solution if more high-precision measurement data is available.
 We have tested our code under controlled conditions as well as in operating oil and gas facilities.
 
-The information on the strength and the approximate location of methane emission sources provided by pyELQ can help 
-operators with more efficient identification and quantification of (unexpected) methane sources, in order to start 
-appropriate mitigating actions accordingly. The pyELQ code is being made available in an open-source environment, 
+The information on the strength and the approximate location of methane emission sources provided by pyELQ can help
+operators with more efficient identification and quantification of (unexpected) methane sources, in order to start
+appropriate mitigating actions accordingly. The pyELQ code is being made available in an open-source environment,
 to support various assets in their quest to reduce methane emissions.
 
 Use cases where the pyELQ code has been applied are described in the following papers:
@@ -81,30 +82,30 @@ Use cases where the pyELQ code has been applied are described in the following p
 * Weidmann, D., Hirst, B. et al. "Locating and Quantifying Methane Emissions by Inverse Analysis of Path-Integrated Concentration Data Using a Markov-Chain Monte Carlo Approach." ACS Earth and Space Chemistry 2022 6 (9), 2190-2198  (https://doi.org/10.1021/acsearthspacechem.2c00093)
 
 ## Deployment design
-The pyELQ code needs high-quality methane concentration and wind data to be able to provide reliable output on location 
-and quantification of methane emission sources. This requires methane concentration sensors of sufficiently high 
-precision in a layout that allows the detection of relevant methane emission sources, in combination with wind 
-measurements of high enough frequency and accuracy. The optimal sensor layout typically depends on the prevailing 
-meteorological conditions at the site of interest and requires multiple concentration sensors to cover the site under 
-different wind directions. 
+The pyELQ code needs high-quality methane concentration and wind data to be able to provide reliable output on location
+and quantification of methane emission sources. This requires methane concentration sensors of sufficiently high
+precision in a layout that allows the detection of relevant methane emission sources, in combination with wind
+measurements of high enough frequency and accuracy. The optimal sensor layout typically depends on the prevailing
+meteorological conditions at the site of interest and requires multiple concentration sensors to cover the site under
+different wind directions.
 
 ## pyELQ data interpretation
-The results from pyELQ come with uncertainty ranges that are representative of probability density functions sampled 
-by a Markov Chain Monte Carlo method. One should take these uncertainty ranges into account when interpreting the pyELQ 
-output data. Remember that absence of evidence for methane emissions does not always imply evidence for absence of 
-methane emissions; for instance, when meteorological conditions are such that there is no sensor downwind of a methane 
-source during the selected monitoring period, then it will be impossible to detect, localize and quantify 
-this particular source. 
-Also, there are limitations to the forward dispersion model which is used in the analysis. 
-For example, the performance of the Gaussian plume dispersion model will degrade at lower wind speeds. 
-Therefore, careful interpretation of the data is always required. 
+The results from pyELQ come with uncertainty ranges that are representative of probability density functions sampled
+by a Markov Chain Monte Carlo method. One should take these uncertainty ranges into account when interpreting the pyELQ
+output data. Remember that absence of evidence for methane emissions does not always imply evidence for absence of
+methane emissions; for instance, when meteorological conditions are such that there is no sensor downwind of a methane
+source during the selected monitoring period, then it will be impossible to detect, localize and quantify
+this particular source.
+Also, there are limitations to the forward dispersion model which is used in the analysis.
+For example, the performance of the Gaussian plume dispersion model will degrade at lower wind speeds.
+Therefore, careful interpretation of the data is always required.
 
 ***
 # Installing pyELQ as a package
 Suppose you want to use this pyELQ package in a different project.
-You can install it from [PyPi](https://pypi.org/project/pyelq/) through pip 
+You can install it from [PyPi](https://pypi.org/project/pyelq/) through pip
 `pip install pyelq`.
-Or you could clone the repository and install it from the source code. 
+Or you could clone the repository and install it from the source code.
 After activating the environment you want to install pyELQ in, open a terminal, move to the main pyELQ folder
 where pyproject.toml is located and run `pip install .`, optionally you can pass the `-e` flag is for editable mode.
 All the main options, info and settings for the package are found in the pyproject.toml file which sits in this repo

{pyelq-1.1.3 → pyelq-1.2.0}/README.md

@@ -22,27 +22,27 @@ SPDX-License-Identifier: Apache-2.0
 </div>
 
 # pyELQ
-This repository contains the Python Emission Localization and Quantification software we call pyELQ. It is code used 
-for gas dispersion modelling, in particular methane emissions detection, localization and quantification. 
+This repository contains the Python Emission Localization and Quantification software we call pyELQ. It is code used
+for gas dispersion modelling, in particular methane emissions detection, localization and quantification.
 
 ***
 # Background
-The **py**thon **E**mission **L**ocalization and **Q**uantification (pyELQ) code aims to maximize effective use of 
-existing measurement data, especially from continuous monitoring solutions. The code has been developed to detect, 
-localize, and quantify methane emissions from concentration and wind measurements. It can be used in combination with 
+The **py**thon **E**mission **L**ocalization and **Q**uantification (pyELQ) code aims to maximize effective use of
+existing measurement data, especially from continuous monitoring solutions. The code has been developed to detect,
+localize, and quantify methane emissions from concentration and wind measurements. It can be used in combination with
 point or beam sensors that are placed strategically on an area of interest.
 
-The algorithms in the pyELQ code are based a Bayesian statistics framework. pyELQ can ingest long-term concentration 
-and wind data, and it performs an inversion to predict the likely strengths and locations of persistent methane sources. 
-The goal is to arrive at a plausible estimate of methane emissions from an area of interest that matches the measured 
-data. The predictions from pyELQ come with uncertainty ranges that are representative of probability density functions 
-sampled by a Markov Chain Monte Carlo method. Time series of varying length can be processed by pyELQ: in general, 
-the Bayesian inversion leads to a more constrained solution if more high-precision measurement data is available. 
+The algorithms in the pyELQ code are based a Bayesian statistics framework. pyELQ can ingest long-term concentration
+and wind data, and it performs an inversion to predict the likely strengths and locations of persistent methane sources.
+The goal is to arrive at a plausible estimate of methane emissions from an area of interest that matches the measured
+data. The predictions from pyELQ come with uncertainty ranges that are representative of probability density functions
+sampled by a Markov Chain Monte Carlo method. Time series of varying length can be processed by pyELQ: in general,
+the Bayesian inversion leads to a more constrained solution if more high-precision measurement data is available.
 We have tested our code under controlled conditions as well as in operating oil and gas facilities.
 
-The information on the strength and the approximate location of methane emission sources provided by pyELQ can help 
-operators with more efficient identification and quantification of (unexpected) methane sources, in order to start 
-appropriate mitigating actions accordingly. The pyELQ code is being made available in an open-source environment, 
+The information on the strength and the approximate location of methane emission sources provided by pyELQ can help
+operators with more efficient identification and quantification of (unexpected) methane sources, in order to start
+appropriate mitigating actions accordingly. The pyELQ code is being made available in an open-source environment,
 to support various assets in their quest to reduce methane emissions.
 
 Use cases where the pyELQ code has been applied are described in the following papers:
@@ -52,30 +52,30 @@ Use cases where the pyELQ code has been applied are described in the following p
 * Weidmann, D., Hirst, B. et al. "Locating and Quantifying Methane Emissions by Inverse Analysis of Path-Integrated Concentration Data Using a Markov-Chain Monte Carlo Approach." ACS Earth and Space Chemistry 2022 6 (9), 2190-2198  (https://doi.org/10.1021/acsearthspacechem.2c00093)
 
 ## Deployment design
-The pyELQ code needs high-quality methane concentration and wind data to be able to provide reliable output on location 
-and quantification of methane emission sources. This requires methane concentration sensors of sufficiently high 
-precision in a layout that allows the detection of relevant methane emission sources, in combination with wind 
-measurements of high enough frequency and accuracy. The optimal sensor layout typically depends on the prevailing 
-meteorological conditions at the site of interest and requires multiple concentration sensors to cover the site under 
-different wind directions. 
+The pyELQ code needs high-quality methane concentration and wind data to be able to provide reliable output on location
+and quantification of methane emission sources. This requires methane concentration sensors of sufficiently high
+precision in a layout that allows the detection of relevant methane emission sources, in combination with wind
+measurements of high enough frequency and accuracy. The optimal sensor layout typically depends on the prevailing
+meteorological conditions at the site of interest and requires multiple concentration sensors to cover the site under
+different wind directions.
 
 ## pyELQ data interpretation
-The results from pyELQ come with uncertainty ranges that are representative of probability density functions sampled 
-by a Markov Chain Monte Carlo method. One should take these uncertainty ranges into account when interpreting the pyELQ 
-output data. Remember that absence of evidence for methane emissions does not always imply evidence for absence of 
-methane emissions; for instance, when meteorological conditions are such that there is no sensor downwind of a methane 
-source during the selected monitoring period, then it will be impossible to detect, localize and quantify 
-this particular source. 
-Also, there are limitations to the forward dispersion model which is used in the analysis. 
-For example, the performance of the Gaussian plume dispersion model will degrade at lower wind speeds. 
-Therefore, careful interpretation of the data is always required. 
+The results from pyELQ come with uncertainty ranges that are representative of probability density functions sampled
+by a Markov Chain Monte Carlo method. One should take these uncertainty ranges into account when interpreting the pyELQ
+output data. Remember that absence of evidence for methane emissions does not always imply evidence for absence of
+methane emissions; for instance, when meteorological conditions are such that there is no sensor downwind of a methane
+source during the selected monitoring period, then it will be impossible to detect, localize and quantify
+this particular source.
+Also, there are limitations to the forward dispersion model which is used in the analysis.
+For example, the performance of the Gaussian plume dispersion model will degrade at lower wind speeds.
+Therefore, careful interpretation of the data is always required.
 
 ***
 # Installing pyELQ as a package
 Suppose you want to use this pyELQ package in a different project.
-You can install it from [PyPi](https://pypi.org/project/pyelq/) through pip 
+You can install it from [PyPi](https://pypi.org/project/pyelq/) through pip
 `pip install pyelq`.
-Or you could clone the repository and install it from the source code. 
+Or you could clone the repository and install it from the source code.
 After activating the environment you want to install pyELQ in, open a terminal, move to the main pyELQ folder
 where pyproject.toml is located and run `pip install .`, optionally you can pass the `-e` flag is for editable mode.
 All the main options, info and settings for the package are found in the pyproject.toml file which sits in this repo

{pyelq-1.1.3 → pyelq-1.2.0}/pyproject.toml

@@ -8,7 +8,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "pyelq"
-version = "1.1.3"
+version = "1.2.0"
 description = "Package for detection, localization and quantification code."
 authors = ["Bas van de Kerkhof", "Matthew Jones", "David Randell"]
 homepage = "https://sede-open.github.io/pyELQ/"
@@ -20,33 +20,33 @@ keywords = ["gas dispersion", "emission", "detection", "localization", "quantifi
 packages = [{ include = "pyelq", from = "src" }]
 
 [tool.poetry.dependencies]
-python = ">=3.9"
-pandas = ">=2.2.3"
-numpy = ">=2.0.2"
-plotly = ">=6.0.0"
-scipy = ">=1.13.1"
-pymap3d = ">=3.1.0"
+python = ">=3.11"
+pandas = ">=2.3.3, <3.0.0"
+numpy = ">=2.4.0"
+plotly = ">=6.5.0"
+scipy = ">=1.16.3"
+pymap3d = ">=3.2.0"
 geojson = ">=3.2.0"
-shapely = ">=2.0.7"
-scikit-learn = ">=1.6.1"
-openmcmc = "==1.0.6"
+shapely = ">=2.1.2"
+scikit-learn = ">=1.8.0"
+openmcmc = "==1.0.8"
 
 [tool.poetry.group.contributor]
 optional = true
 
 [tool.poetry.group.contributor.dependencies]
-black = ">=25.1.0"
-isort = ">=6.0.0"
+black = ">=25.12.0"
+isort = ">=7.0.0"
 pydocstyle = ">=6.3.0"
-pylint = ">=3.3.4"
-pytest = ">=8.3.4"
-pytest-cov = ">=6.0.0"
-mkdocs-material = ">=9.6.2"
-mkdocstrings-python = ">=1.14.5"
+pylint = ">=4.0.4"
+pytest = ">=9.0.2"
+pytest-cov = ">=7.0.0"
+mkdocs-material = ">=9.7.1"
+mkdocstrings-python = ">=2.0.1"
 
 
 [tool.poetry.group.dev.dependencies]
-ipykernel = "^6.29.5"
+ipykernel = "^7.1.0"
 
 [tool.pytest.ini_options]
 addopts = "--cov=pyelq --cov-fail-under=90 --ignore-glob=*plot*"

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/__init__.py

@@ -2,6 +2,7 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 """Main pyELQ module."""
+
 __all__ = [
     "component",
     "data_access",

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/component/__init__.py

@@ -3,4 +3,5 @@
 # SPDX-License-Identifier: Apache-2.0
 
 """Components Module."""
+
 __all__ = ["background", "component", "error_model", "offset", "source_model"]

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/component/background.py

@@ -1,7 +1,6 @@
 # SPDX-FileCopyrightText: 2024 Shell Global Solutions International B.V. All Rights Reserved.
 #
 # SPDX-License-Identifier: Apache-2.0
-
 """Model components for background modelling."""
 
 from abc import abstractmethod
@@ -22,7 +21,7 @@ from sklearn.neighbors import NearestNeighbors
 from pyelq.component.component import Component
 from pyelq.coordinate_system import Coordinate
 from pyelq.gas_species import GasSpecies
-from pyelq.meteorology import MeteorologyGroup
+from pyelq.meteorology.meteorology import MeteorologyGroup
 from pyelq.sensor.beam import Beam
 from pyelq.sensor.sensor import SensorGroup
 
@@ -51,6 +50,7 @@ class Background(Component):
         initial_precision (float): initial value for the scalar precision parameter.
         basis_matrix (sparse.csr_array): [n_obs x n_time] matrix mapping the background model parameters on to the
             observations.
+        precision_time_0 (float): precision relating to the first time stamp in the model. Defaults to 0.01.
 
     """
 
@@ -65,6 +65,7 @@ class Background(Component):
     prior_precision_rate: float = 1e-3
     initial_precision: float = 1.0
     basis_matrix: sparse.csr_array = field(init=False)
+    precision_time_0: float = field(init=False, default=0.01)
 
     @abstractmethod
     def initialise(self, sensor_object: SensorGroup, meteorology: MeteorologyGroup, gas_species: GasSpecies):
@@ -168,6 +169,9 @@ class TemporalBackground(Background):
     def initialise(self, sensor_object: SensorGroup, meteorology: MeteorologyGroup, gas_species: GasSpecies):
         """Create temporal background model from sensor, meteorology and gas species inputs.
 
+        The precision matrix is made to be full rank by adjusting the precision at the first time point using the
+        precision_time_0 attribute.
+
         Args:
             sensor_object (SensorGroup): sensor data object.
             meteorology (MeteorologyGroup): meteorology data object.
@@ -180,6 +184,8 @@ class TemporalBackground(Background):
         self.n_parameter = len(self.time)
         self.basis_matrix = sparse.csr_array((np.ones(self.n_obs), (np.array(range(self.n_obs)), unique_inverse)))
         self.precision_matrix = gmrf.precision_temporal(time=self.time)
+        lam = self.precision_matrix[0, 0]
+        self.precision_matrix[0, 0] = lam * (2.0 - lam / (self.precision_time_0 + lam))
         if self.mean_bg is None:
             self.mean_bg = gas_species.global_background
 
@@ -193,8 +199,8 @@ class SpatioTemporalBackground(Background):
     The background parameter is an [n_location * n_time x 1] (if self.spatial_dependence is True) or an [n_time x 1]
     vector (if self.spatial_dependence is False). In the spatio-temporal case, the background vector is assumed to
     unwrap over space and time as follows:
-    bg = [b_1(t_1), b_2(t_1),..., b_nlct(t_1),...,b_1(t_k),..., b_nlct(t_k),...].T
-    where nlct is the number of sensor locations.
+    bg = [b_1(t_1), b_2(t_1),..., b_n_lct(t_1),...,b_1(t_k),..., b_n_lct(t_k),...].T
+    where n_lct is the number of sensor locations.
     This unwrapping mechanism is chosen as it greatly speeds up the sparse matrix operations in the solver (vs. the
     alternative).
 
@@ -226,7 +232,6 @@ class SpatioTemporalBackground(Background):
         spatial_precision_matrix (np.ndarray): spatial component of the precision matrix. The full model precision
             matrix is the Kronecker product of this matrix with the self.temporal_precision_matrix. Simply set to 1 if
             self.spatial_dependence is False.
-        precision_time_0 (float): precision relating to the first time stamp in the model. Defaults to 0.01.
 
     """
 
@@ -238,7 +243,6 @@ class SpatioTemporalBackground(Background):
     location: Coordinate = field(init=False)
     temporal_precision_matrix: Union[np.ndarray, sparse.csc_matrix] = field(init=False)
     spatial_precision_matrix: np.ndarray = field(init=False)
-    precision_time_0: float = field(init=False, default=0.01)
 
     def initialise(self, sensor_object: SensorGroup, meteorology: MeteorologyGroup, gas_species: GasSpecies):
         """Take data inputs and extract relevant properties.
@@ -303,10 +307,9 @@ class SpatioTemporalBackground(Background):
             self.time = pd.array(np.unique(sensor_object.time), dtype="datetime64[ns]")
             self.n_time = len(self.time)
         else:
-            self.time = pd.array(
-                pd.date_range(start=np.min(sensor_object.time), end=np.max(sensor_object.time), periods=self.n_time),
-                dtype="datetime64[ns]",
-            )
+            self.time = pd.date_range(
+                start=np.min(sensor_object.time), end=np.max(sensor_object.time), periods=self.n_time
+            ).array
 
     def make_spatial_knots(self, sensor_object: SensorGroup):
         """Create the spatial grid for the model.
@@ -333,6 +336,9 @@ class SpatioTemporalBackground(Background):
 
         Defined as the Kronecker product of the temporal precision matrix and the spatial precision matrix.
 
+        The precision matrix is made to be full rank by adjusting the precision at the first time point using the
+        precision_time_0 attribute.
+
         """
         self.temporal_precision_matrix = gmrf.precision_temporal(time=self.time)
         lam = self.temporal_precision_matrix[0, 0]
@@ -386,6 +392,6 @@ class SpatioTemporalBackground(Background):
                 self.location.north[k] = np.mean(sensor.location.to_enu().north, axis=0)
                 self.location.up[k] = np.mean(sensor.location.to_enu().up, axis=0)
             else:
-                self.location.east[k] = sensor.location.to_enu().east
-                self.location.north[k] = sensor.location.to_enu().north
-                self.location.up[k] = sensor.location.to_enu().up
+                self.location.east[k] = sensor.location.to_enu().east.item()
+                self.location.north[k] = sensor.location.to_enu().north.item()
+                self.location.up[k] = sensor.location.to_enu().up.item()

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/component/component.py

@@ -3,13 +3,14 @@
 # SPDX-License-Identifier: Apache-2.0
 
 """Superclass for model components."""
+
 from abc import ABC, abstractmethod
 from dataclasses import dataclass
 
 from openmcmc.model import Model
 
 from pyelq.gas_species import GasSpecies
-from pyelq.meteorology import MeteorologyGroup
+from pyelq.meteorology.meteorology import MeteorologyGroup
 from pyelq.sensor.sensor import SensorGroup
 
 

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/component/error_model.py

@@ -4,6 +4,7 @@
 
 # -*- coding: utf-8 -*-
 """Error model module."""
+
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING, Union
 
@@ -15,7 +16,7 @@ from openmcmc.sampler.sampler import NormalGamma
 
 from pyelq.component.component import Component
 from pyelq.gas_species import GasSpecies
-from pyelq.meteorology import MeteorologyGroup
+from pyelq.meteorology.meteorology import MeteorologyGroup
 from pyelq.sensor.sensor import Sensor, SensorGroup
 
 if TYPE_CHECKING:

{pyelq-1.1.3 → pyelq-1.2.0}/src/pyelq/component/offset.py

@@ -4,6 +4,7 @@
 
 # -*- coding: utf-8 -*-
 """Offset module."""
+
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING, Union
 
@@ -17,7 +18,7 @@ from scipy import sparse
 
 from pyelq.component.component import Component
 from pyelq.gas_species import GasSpecies
-from pyelq.meteorology import Meteorology
+from pyelq.meteorology.meteorology import Meteorology
 from pyelq.sensor.sensor import Sensor, SensorGroup
 
 if TYPE_CHECKING: