mt-metadata 0.3.9__tar.gz → 0.4.0__tar.gz

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/MANIFEST.in

@@ -18,8 +18,12 @@ recursive-include mt_metadata/transfer_functions/io/edi/metadata/standards *.jso
 recursive-include mt_metadata/transfer_functions/io/zfiles/metadata/standards *.json
 recursive-include mt_metadata/transfer_functions/io/jfiles/metadata/standards *.json
 recursive-include mt_metadata/transfer_functions/io/zonge/metadata/standards *.json
+recursive-include mt_metadata/transfer_functions/processing/standards *.json
 recursive-include mt_metadata/transfer_functions/processing/aurora/standards *.json
 recursive-include mt_metadata/transfer_functions/processing/fourier_coefficients/standards *.json
+recursive-include mt_metadata/features/standards *.json
+recursive-include mt_metadata/features/test_helpers *.json
+recursive-include mt_metadata/features/weights/standards *.json
 recursive-include mt_metadata/data/mt_xml *.xml
 recursive-include mt_metadata/data/stationxml *.xml
 recursive-include mt_metadata/data/transfer_functions *.edi *.xml *.j *.zmm *.zss *.avg

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: mt_metadata
-Version: 0.3.9
+Version: 0.4.0
 Summary: Metadata for magnetotelluric data
 Home-page: https://github.com/kujaku11/mt_metadata
 Author: Jared Peacock
@@ -21,8 +21,30 @@ Requires-Python: >=3.7
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: AUTHORS.rst
-
-# mt_metadata version 0.3.9
+Requires-Dist: numpy
+Requires-Dist: scipy
+Requires-Dist: pandas
+Requires-Dist: matplotlib
+Requires-Dist: xarray
+Requires-Dist: loguru
+Provides-Extra: obspy
+Requires-Dist: obspy; extra == "obspy"
+Provides-Extra: test
+Requires-Dist: pytest>=3; extra == "test"
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: license
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# mt_metadata version 0.4.0
  Standard MT metadata
 
 [![PyPi version](https://img.shields.io/pypi/v/mt_metadata.svg)](https://pypi.python.org/pypi/mt-metadata)
@@ -37,11 +59,11 @@ License-File: AUTHORS.rst
 
 MT Metadata is a project led by [IRIS-PASSCAL MT Software working group](https://www.iris.edu/hq/about_iris/governance/mt_soft>) and USGS to develop tools that standardize magnetotelluric metadata, well, at least create tools for standards that are generally accepted.  This include the two main types of magnetotelluric data
 
-- **Time Series** 
+- **Time Series**
     - Structured as:
         - Experiment -> Survey -> Station -> Run -> Channel
     - Supports translation to/from **StationXML**
-        
+
 - **Transfer Functions**
     - Supports (will support) to/from:
         - **EDI** (most common format)
@@ -52,11 +74,12 @@ MT Metadata is a project led by [IRIS-PASSCAL MT Software working group](https:/
 
 Most people will be using the transfer functions, but a lot of that metadata comes from the time series metadata.  This module supports both and has tried to make them more or less seamless to reduce complication.
 
-* **Version**: 0.3.9
+* **Version**: 0.4.0
 * **Free software**: MIT license
 * **Documentation**: https://mt-metadata.readthedocs.io.
 * **Examples**: Click the `Binder` badge above and Jupyter Notebook examples are in **mt_metadata/examples/notebooks** and **docs/source/notebooks**
 * **Suggested Citation**: Peacock, J. R., Kappler, K., Ronan, T., Heagy, L.,  Kelbert, A., Frassetto, A. (2022) MTH5: An archive and exchangeable data format for magnetotelluric time series data, *Computers & Geoscience*, **162**, doi:10.1016/j.cageo.2022.105102
+* **IPDS**: IP-138156
 
 
 # Installation
@@ -65,20 +88,27 @@ Most people will be using the transfer functions, but a lot of that metadata com
 
 `git clone https://github.com/kujaku11/mt_metadata.git`
 
-`python setup.py install`
+`pip install .`
 
-You can add the flag `-e` if you want to change the code.
+You can add the flag `-e` if you want to install the source repository in an editable state.
 
 ## PIP
 `pip install mt_metadata`
 
+> You can install with optional packages by appending `[option_name]` to the package name during the
+> `pip` install command. E.g:
+>
+> `pip install mt_metadata[obspy]`
+>
+> or `pip install .[obspy]` if building from source.
+
 ## Conda
 
 `conda install mt_metadata`
 
 # Standards
 
-Each metadata keyword has an associated standard that goes with it.  These are stored internally in JSON file.  The JSON files are read in when the package is loaded to initialize the standards.  Each keyword is described by:  
+Each metadata keyword has an associated standard that goes with it.  These are stored internally in JSON file.  The JSON files are read in when the package is loaded to initialize the standards.  Each keyword is described by:
 
 - **type** - How the value should be represented based on very basic types
 
@@ -94,7 +124,7 @@ Each metadata keyword has an associated standard that goes with it.  These are s
     - *Controlled Vocabulary* only certain values are allowed according to **options**
     - *Date* a date and/or time string in ISO format
     - *Number* a float or integer
-    - *Boolean* the value can only be True or False 
+    - *Boolean* the value can only be True or False
 
 - **units** - Units of the value
 - **description** - Full description of what the metadata key is meant to convey.
@@ -111,7 +141,7 @@ Each metadata object is based on a Base class that has methods:
 - to_from_dict
 - attribute_information
 
-And each object has a doc string that describes the standard: 
+And each object has a doc string that describes the standard:
 
 
 | **Metadata Key**                             | **Description**                               | **Example**    |
@@ -224,7 +254,7 @@ print(x.to_xml(string=True))
 Credits
 -------
 
-This project is in cooperation with the Incorporated Research Institutes of Seismology, the U.S. Geological Survey, and other collaborators.  Facilities of the IRIS Consortium are supported by the National Science Foundation’s Seismological Facilities for the Advancement of Geoscience (SAGE) Award under Cooperative Support Agreement EAR-1851048.  USGS is partially funded through the Community for Data Integration and IMAGe through the Minerals Resources Program. 
+This project is in cooperation with the Incorporated Research Institutes of Seismology, the U.S. Geological Survey, and other collaborators.  Facilities of the IRIS Consortium are supported by the National Science Foundation’s Seismological Facilities for the Advancement of Geoscience (SAGE) Award under Cooperative Support Agreement EAR-1851048.  USGS is partially funded through the Community for Data Integration and IMAGe through the Minerals Resources Program.
 
 
 History

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/README.md

@@ -1,4 +1,4 @@
-# mt_metadata version 0.3.9
+# mt_metadata version 0.4.0
  Standard MT metadata
 
 [![PyPi version](https://img.shields.io/pypi/v/mt_metadata.svg)](https://pypi.python.org/pypi/mt-metadata)
@@ -13,11 +13,11 @@
 
 MT Metadata is a project led by [IRIS-PASSCAL MT Software working group](https://www.iris.edu/hq/about_iris/governance/mt_soft>) and USGS to develop tools that standardize magnetotelluric metadata, well, at least create tools for standards that are generally accepted.  This include the two main types of magnetotelluric data
 
-- **Time Series** 
+- **Time Series**
     - Structured as:
         - Experiment -> Survey -> Station -> Run -> Channel
     - Supports translation to/from **StationXML**
-        
+
 - **Transfer Functions**
     - Supports (will support) to/from:
         - **EDI** (most common format)
@@ -28,11 +28,12 @@ MT Metadata is a project led by [IRIS-PASSCAL MT Software working group](https:/
 
 Most people will be using the transfer functions, but a lot of that metadata comes from the time series metadata.  This module supports both and has tried to make them more or less seamless to reduce complication.
 
-* **Version**: 0.3.9
+* **Version**: 0.4.0
 * **Free software**: MIT license
 * **Documentation**: https://mt-metadata.readthedocs.io.
 * **Examples**: Click the `Binder` badge above and Jupyter Notebook examples are in **mt_metadata/examples/notebooks** and **docs/source/notebooks**
 * **Suggested Citation**: Peacock, J. R., Kappler, K., Ronan, T., Heagy, L.,  Kelbert, A., Frassetto, A. (2022) MTH5: An archive and exchangeable data format for magnetotelluric time series data, *Computers & Geoscience*, **162**, doi:10.1016/j.cageo.2022.105102
+* **IPDS**: IP-138156
 
 
 # Installation
@@ -41,20 +42,27 @@ Most people will be using the transfer functions, but a lot of that metadata com
 
 `git clone https://github.com/kujaku11/mt_metadata.git`
 
-`python setup.py install`
+`pip install .`
 
-You can add the flag `-e` if you want to change the code.
+You can add the flag `-e` if you want to install the source repository in an editable state.
 
 ## PIP
 `pip install mt_metadata`
 
+> You can install with optional packages by appending `[option_name]` to the package name during the
+> `pip` install command. E.g:
+>
+> `pip install mt_metadata[obspy]`
+>
+> or `pip install .[obspy]` if building from source.
+
 ## Conda
 
 `conda install mt_metadata`
 
 # Standards
 
-Each metadata keyword has an associated standard that goes with it.  These are stored internally in JSON file.  The JSON files are read in when the package is loaded to initialize the standards.  Each keyword is described by:  
+Each metadata keyword has an associated standard that goes with it.  These are stored internally in JSON file.  The JSON files are read in when the package is loaded to initialize the standards.  Each keyword is described by:
 
 - **type** - How the value should be represented based on very basic types
 
@@ -70,7 +78,7 @@ Each metadata keyword has an associated standard that goes with it.  These are s
     - *Controlled Vocabulary* only certain values are allowed according to **options**
     - *Date* a date and/or time string in ISO format
     - *Number* a float or integer
-    - *Boolean* the value can only be True or False 
+    - *Boolean* the value can only be True or False
 
 - **units** - Units of the value
 - **description** - Full description of what the metadata key is meant to convey.
@@ -87,7 +95,7 @@ Each metadata object is based on a Base class that has methods:
 - to_from_dict
 - attribute_information
 
-And each object has a doc string that describes the standard: 
+And each object has a doc string that describes the standard:
 
 
 | **Metadata Key**                             | **Description**                               | **Example**    |
@@ -200,4 +208,4 @@ print(x.to_xml(string=True))
 Credits
 -------
 
-This project is in cooperation with the Incorporated Research Institutes of Seismology, the U.S. Geological Survey, and other collaborators.  Facilities of the IRIS Consortium are supported by the National Science Foundation’s Seismological Facilities for the Advancement of Geoscience (SAGE) Award under Cooperative Support Agreement EAR-1851048.  USGS is partially funded through the Community for Data Integration and IMAGe through the Minerals Resources Program. 
+This project is in cooperation with the Incorporated Research Institutes of Seismology, the U.S. Geological Survey, and other collaborators.  Facilities of the IRIS Consortium are supported by the National Science Foundation’s Seismological Facilities for the Advancement of Geoscience (SAGE) Award under Cooperative Support Agreement EAR-1851048.  USGS is partially funded through the Community for Data Integration and IMAGe through the Minerals Resources Program.

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/mt_metadata/__init__.py

@@ -39,7 +39,7 @@ you should only have to changes these dictionaries.
 
 __author__ = """Jared Peacock"""
 __email__ = "jpeacock@usgs.gov"
-__version__ = "0.3.9"
+__version__ = "0.4.0"
 
 # =============================================================================
 # Imports

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/mt_metadata/base/helpers.py

@@ -252,9 +252,7 @@ def write_block(key, attr_dict, c1=45, c2=45, c3=15):
         f"       :widths: {c1} {c2} {c3}",
         "",
         hline,
-        line.format(
-            f"**{key}**", c1, "**Description**", c2, "**Example**", c3
-        ),
+        line.format(f"**{key}**", c1, "**Description**", c2, "**Example**", c3),
         mline,
     ]
 
@@ -449,14 +447,43 @@ def recursive_split_getattr(base_object, name, sep="."):
     return value, prop
 
 
-def recursive_split_setattr(base_object, name, value, sep="."):
+def recursive_split_setattr(
+    base_object, name, value, sep=".", skip_validation=False
+):
+    """
+    Recursively split a name and set the value of the last key. Recursion splits on the separator present in the name.
+
+    :param base_object: The object having its attribute set, or a "parent" object in the recursive/nested scenario
+    :type base_object: object
+    :param name: The name of the attribute to set
+    :type name: str
+    :param value: The value to set the attribute to
+    :type value: any
+    :param sep: The separator to split the name on, defaults to "."
+    :type sep: str, optional
+    :param skip_validation: Whether to skip validation/parse of the attribute, defaults to False
+    :type skip_validation: Optional[bool]
+
+    :return: None
+    :rtype: NoneType
+
+    """
     key, *other = name.split(sep, 1)
 
-    if other:
-        base_object = getattr(base_object, key)
-        recursive_split_setattr(base_object, other[0], value)
+    if skip_validation:
+        if other:
+            base_object = getattr(base_object, key)
+            recursive_split_setattr(
+                base_object, other[0], value, skip_validation=True
+            )
+        else:
+            base_object.setattr_skip_validation(key, value)
     else:
-        setattr(base_object, key, value)
+        if other:
+            base_object = getattr(base_object, key)
+            recursive_split_setattr(base_object, other[0], value)
+        else:
+            setattr(base_object, key, value)
 
 
 def structure_dict(meta_dict, sep="."):
@@ -637,7 +664,6 @@ def element_to_string(element):
 # Helper function to be sure everything is encoded properly
 # =============================================================================
 class NumpyEncoder(json.JSONEncoder):
-
     """
     Need to encode numpy ints and floats for json to work
     """
@@ -694,3 +720,52 @@ def validate_name(name, pattern=None):
     if name is None:
         return "unknown"
     return name.replace(" ", "_")
+
+
+def requires(**requirements):
+    """Decorate a function with optional dependencies.
+
+    Parameters
+    ----------
+    **requirements : obj
+        keywords of package name and the required object for
+        a function.
+
+    Returns
+    -------
+    decorated_function : function
+        Original function if all soft dependencies are met, otherwise
+        it returns an empty function which prints why it is not running.
+
+    Examples
+    --------
+    ```
+    try:
+        import obspy
+    except ImportError:
+        obspy = None
+
+    @requires(obspy=obspy)
+    def obspy_function():
+        ...
+        # does something using obspy
+
+    """
+    # Check the requirements, add missing package name in the list `missing`.
+    missing = []
+    for key, item in requirements.items():
+        if not item:
+            missing.append(key)
+
+    def decorated_function(function):
+        """Wrap function."""
+        if not missing:
+            return function
+        else:
+            def passer(*args, **kwargs):
+                print(("Missing dependencies: {d}.".format(d=missing)))
+                print(("Not running `{}`.".format(function.__name__)))
+
+            return passer
+
+    return decorated_function

{mt_metadata-0.3.9 → mt_metadata-0.4.0}/mt_metadata/base/metadata.py

@@ -16,6 +16,7 @@ from collections import OrderedDict
 from operator import itemgetter
 from pathlib import Path
 from loguru import logger
+from typing import Optional, Union
 
 import json
 import pandas as pd
@@ -31,6 +32,8 @@ from . import helpers
 from mt_metadata.base.helpers import write_lines
 
 attr_dict = {}
+
+
 # =============================================================================
 #  Base class that everything else will inherit
 # =============================================================================
@@ -38,9 +41,9 @@ attr_dict = {}
 
 class Base:
     __doc__ = write_lines(attr_dict)
+    _base_attr_loaded = False
 
     def __init__(self, attr_dict={}, **kwargs):
-
         self._changed = False
 
         self._class_name = validate_attribute(self.__class__.__name__)
@@ -48,12 +51,17 @@ class Base:
         self.logger = logger
         self._debug = False
 
-        self._set_attr_dict(attr_dict)
+        # attr_dict from subclass has already been validated on json load, so
+        # we shouldn't need to validate it again re-validation of the attribute
+        # dictionary used to contribute to some slowness in instantiation of
+        # subclasses
+        # self._set_attr_dict(deepcopy(attr_dict), skip_validation=True)
+        self._set_attr_dict(attr_dict, skip_validation=True)
 
         for name, value in kwargs.items():
-            self.set_attr_from_name(name, value)
+            self.set_attr_from_name(name, value, skip_validation=False)
 
-    def _set_attr_dict(self, attr_dict):
+    def _set_attr_dict(self, attr_dict, skip_validation=False):
         """
         Set attribute dictionary and variables.
 
@@ -61,13 +69,15 @@ class Base:
 
         :param attr_dict: attribute dictionary
         :type attr_dict: dict
+        :param skip_validation: skip validation/parse of the attribute dictionary
+        :type skip_validation: bool
 
         """
 
         self._attr_dict = attr_dict
 
         for key, value_dict in attr_dict.items():
-            self.set_attr_from_name(key, value_dict["default"])
+            self.set_attr_from_name(key, value_dict["default"], skip_validation)
 
     def __str__(self):
         """
@@ -85,62 +95,99 @@ class Base:
     def __repr__(self):
         return self.to_json()
 
-    def __eq__(self, other):
-        if other in [None]:
+    def __eq__(
+        self,
+        other: Union["Base", dict, str, pd.Series],
+        ignore_keys: Optional[Union[list, tuple]] = None
+    ) -> bool:
+        """
+
+            Checks for equality between self and input argument `other`.
+
+            Logic:
+             - verify that object is of expected dtype (else return False)
+             - form two dicts, one representing self and one representing other
+             - compare the two dicts
+             - return the outcome of the comparison
+
+            :param other: Another Base object, or it's representation as dict, str, pd.Series
+            :type other: Union["Base", dict, str, pd.Series]
+            :param ignore_keys: An iterable of keys to ignore during the comparison.
+            :type other: Union["Base", dict, str, pd.Series]
+
+            TODO: Once python 3.10 and lower are supported, change "Base" to Self in typehints.
+
+        """
+        # validate dtype
+        allowed_input_dtypes = (Base, dict, str, pd.Series)
+        if not isinstance(other, allowed_input_dtypes):
+            msg = f"Unable to evaluate equality of {type(self)} with {type(other)}"
+            self.logger.info(msg)
             return False
-        elif isinstance(other, (Base, dict, str, pd.Series)):
-            home_dict = self.to_dict(single=True, required=False)
-            if isinstance(other, Base):
-                other_dict = other.to_dict(single=True, required=False)
-            elif isinstance(other, dict):
-                other_dict = other
-            elif isinstance(other, str):
-                if other.lower() in ["none", "null", "unknown"]:
-                    return False
-                other_dict = OrderedDict(
-                    sorted(json.loads(other).items(), key=itemgetter(0))
-                )
-            elif isinstance(other, pd.Series):
-                other_dict = OrderedDict(
-                    sorted(other.to_dict().items(), key=itemgetter(0))
-                )
-            else:
-                raise ValueError(
-                    f"Cannot compare {self._class_name} with {type(other)}"
-                )
-            fail = False
-            for key, value in home_dict.items():
-                try:
-                    other_value = other_dict[key]
-                    if isinstance(value, np.ndarray):
-                        if value.size != other_value.size:
-                            msg = f"Array sizes for {key} differ: {value.size} != {other_value.size}"
-                            self.logger.info(msg)
-                            fail=True
-                            continue
-                        if not (value == other_value).all():
-                            msg = f"{key}: {value} != {other_value}"
-                            self.logger.info(msg)
-                            fail = True
-                    elif isinstance(value, (float, int, complex)):
-                        if not np.isclose(value, other_value):
-                            msg = f"{key}: {value} != {other_value}"
-                            self.logger.info(msg)
-                            fail = True
-                    else:
-                        if value != other_value:
-                            msg = f"{key}: {value} != {other_value}"
-                            self.logger.info(msg)
-                            fail = True
-                except KeyError:
-                    msg = "Cannot find {0} in other".format(key)
-                    self.logger.info(msg)
-            if fail:
+
+        # form two dicts, one for self and one for other
+        home_dict = self.to_dict(single=True, required=False)
+        if isinstance(other, Base):
+            other_dict = other.to_dict(single=True, required=False)
+        elif isinstance(other, dict):
+            other_dict = other
+        elif isinstance(other, str):
+            if other.lower() in ["none", "null", "unknown"]:
                 return False
-            else:
-                return True
+            other_dict = OrderedDict(
+                sorted(json.loads(other).items(), key=itemgetter(0))
+            )
+        elif isinstance(other, pd.Series):
+            other_dict = OrderedDict(
+                sorted(other.to_dict().items(), key=itemgetter(0))
+            )
         else:
+            raise ValueError(
+                f"Cannot compare {self._class_name} with {type(other)}"
+            )
+
+        # set ignore keys to empty iterable if it was none
+        if ignore_keys is None:
+            ignore_keys = ()
+
+        # Compare the two dicts
+        fail = False
+        for key, value in home_dict.items():
+            if key in ignore_keys:
+                continue
+            try:
+                other_value = other_dict[key]
+                if isinstance(value, np.ndarray):
+                    if value.size != other_value.size:
+                        msg = f"Array sizes for {key} differ: {value.size} != {other_value.size}"
+                        self.logger.info(msg)
+                        fail=True
+                        continue
+                    if not (value == other_value).all():
+                        msg = f"{key}: {value} != {other_value}"
+                        self.logger.info(msg)
+                        fail = True
+                        # TODO: Add np.allclose here?
+                elif isinstance(value, (float, int, complex)):
+                    if not np.isclose(value, other_value):
+                        msg = f"{key}: {value} != {other_value}"
+                        self.logger.info(msg)
+                        fail = True
+                else:
+                    if value != other_value:
+                        msg = f"{key}: {value} != {other_value}"
+                        self.logger.info(msg)
+                        fail = True
+            except KeyError:
+                msg = f"Cannot find key {key} in other"
+                self.logger.info(msg)
+
+        # return the outcome of the comparison
+        if fail:
             return False
+        else:
+            return True
+
 
     def __ne__(self, other):
         return not self.__eq__(other)
@@ -329,7 +376,11 @@ class Base:
                 + f" are allowed.  Allowing {option_list} to be set to {value}."
             )
             return True, other_possible, msg
-        return False, other_possible, f"Value '{value}' for metadata field '{name}' not found in options list {option_list}"
+        return (
+            False,
+            other_possible,
+            f"Value '{value}' for metadata field '{name}' not found in options list {option_list}",
+        )
 
     def __setattr__(self, name, value):
         """
@@ -383,9 +434,8 @@ class Base:
             try:
                 test_property = getattr(self.__class__, name, None)
                 if isinstance(test_property, property):
-                    self.logger.debug(
-                        f"Identified {name} as property, using fset"
-                    )
+                    msg = f"Identified {name} as property, using fset"
+                    self.logger.debug(msg)
                     test_property.fset(self, value)
                     return
             except AttributeError:
@@ -399,7 +449,9 @@ class Base:
                 # check options
                 if v_dict["style"] == "controlled vocabulary":
                     options = v_dict["options"]
-                    accept, other, msg = self._validate_option(name, value, options)
+                    accept, other, msg = self._validate_option(
+                        name, value, options
+                    )
                     if not accept:
                         self.logger.error(msg.format(value, options))
                         raise MTSchemaError(msg.format(value, options))
@@ -473,7 +525,20 @@ class Base:
             return value
         return self._validate_type(value, v_type)
 
-    def set_attr_from_name(self, name, value):
+    def setattr_skip_validation(self, name, value):
+        """
+        Set attribute without validation
+
+        :param name: name of attribute
+        :type name: string
+
+        :param value: value of the new attribute
+        :type value: described in value_dict
+
+        """
+        self.__dict__[name] = value
+
+    def set_attr_from_name(self, name, value, skip_validation=False):
         """
         Helper function to set attribute from the given name.
 
@@ -487,6 +552,8 @@ class Base:
         :type name: string
         :param value: attribute value
         :type value: type is defined by the attribute name
+        :param skip_validation: skip validation/parse of the key-value pair
+        :type skip_validation: bool
 
         :Example:
 
@@ -497,7 +564,9 @@ class Base:
         """
         if "." in name:
             try:
-                helpers.recursive_split_setattr(self, name, value)
+                helpers.recursive_split_setattr(
+                    self, name, value, skip_validation=skip_validation
+                )
             except AttributeError as error:
                 msg = (
                     "{0} is not in the current standards.  "
@@ -508,7 +577,10 @@ class Base:
                 self.logger.error(msg.format(name))
                 raise AttributeError(error)
         else:
-            setattr(self, name, value)
+            if skip_validation:
+                self.setattr_skip_validation(name, value)
+            else:
+                setattr(self, name, value)
 
     def add_base_attribute(self, name, value, value_dict):
         """

mt_metadata-0.4.0/mt_metadata/features/__init__.py

@@ -0,0 +1,14 @@
+# -*- coding: utf-8 -*-
+
+from .feature_ts_run import FeatureTSRun
+from .feature_fc_run import FeatureFCRun
+from .feature_decimation_channel import FeatureDecimationChannel
+from .feature import Feature
+
+
+__all__ = [
+    "FeatureTSRun",
+    "FeatureFCRun",
+    "Feature",
+    "FeatureDecimationChannel",
+]