Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Deprecate TransferFunctionHeader Class #109

Closed
kkappler opened this issue Sep 28, 2021 · 5 comments
Closed

Deprecate TransferFunctionHeader Class #109

kkappler opened this issue Sep 28, 2021 · 5 comments

Comments

@kkappler
Copy link
Collaborator

kkappler commented Sep 28, 2021 •
edited
Loading

See Issue #18. The TransferFunctionHeader object is not needed as we get this information from the mth5 now as dictionaries.

If we want to eliminate TransferFunctionHeader, we need to change the places where the code uses it to get that data from another object. The TransferFunction object would need to be able to provide:

  1. A list of input channels
  2. A list of output channels
  3. a property that returns num_input_channels and num_output_channels bases on 1,2, above.
  4. local_station_id
  5. TransferFunctionCollection will need to get its "header" from the TransferFunction object
  • Note that TransferFunctionCollection also only needs to access input_channels, output_channels, and local_station_id
  1. Finally, the class should be moved into sandbox before deprecation as it does get used in the one-off function aurora/sandbox/io_helpers/matlab_z_file_reader.py which reads matlab structs, and is also slated for deprecation.

Helpers in coding this:
rgreppy tf_header
yields:

aurora/sandbox/io_helpers/matlab_z_file_reader.py:101:tfc = TransferFunctionCollection(header=tf_obj.tf_header, tf_dict=tf_dict)
aurora/sandbox/.ipynb_checkpoints/plot_helpers-checkpoint.py:45:    ttl_str = tf_obj.tf_header.local_station_id
aurora/sandbox/plot_helpers.py:48:    ttl_str = tf_obj.tf_header.local_station_id
aurora/transfer_function/transfer_function_collection.py:56:        return self.tf_dict[0].tf_header.local_station_id
aurora/transfer_function/transfer_function_collection.py:379:                for out_ch in tf.tf_header.output_channels:
aurora/transfer_function/transfer_function_collection.py:380:                    for inp_ch in tf.tf_header.input_channels:
aurora/transfer_function/transfer_function_collection.py:392:                for i, inp_ch1 in enumerate(tf.tf_header.input_channels):
aurora/transfer_function/transfer_function_collection.py:393:                    for inp_ch2 in tf.tf_header.input_channels[: i + 1]:
aurora/transfer_function/transfer_function_collection.py:407:                for i, out_ch1 in enumerate(tf.tf_header.output_channels):
aurora/transfer_function/transfer_function_collection.py:408:                    for out_ch2 in tf.tf_header.output_channels[: i + 1]:
aurora/transfer_function/transfer_function_collection.py:537:        ttl_str = f"{tf.tf_header.local_station_id} {xy_or_yx} \n{ttl_str}"
aurora/transfer_function/transfer_function_collection.py:549:        # figure_basename = f"synthetic_{tf.tf_header.local_station_id}_{xy_or_yx}.png"
aurora/transfer_function/base.py:66:    def __init__(self, tf_header, frequency_bands, **kwargs):
aurora/transfer_function/base.py:76:        tf_header
aurora/transfer_function/base.py:80:        self.tf_header = tf_header
aurora/transfer_function/base.py:89:        if self.tf_header is not None:
aurora/transfer_function/base.py:138:        if self.tf_header is None:
aurora/transfer_function/base.py:151:                "output_channel": self.tf_header.output_channels,
aurora/transfer_function/base.py:152:                "input_channel": self.tf_header.input_channels,
aurora/transfer_function/base.py:165:                "channel": self.tf_header.output_channels,
aurora/transfer_function/base.py:178:                "input_channel_1": self.tf_header.input_channels,
aurora/transfer_function/base.py:179:                "input_channel_2": self.tf_header.input_channels,
aurora/transfer_function/base.py:192:                "output_channel_1": self.tf_header.output_channels,
aurora/transfer_function/base.py:193:                "output_channel_2": self.tf_header.output_channels,
aurora/transfer_function/base.py:204:                "output_channel": self.tf_header.output_channels,
aurora/transfer_function/base.py:221:        return self.tf_header.num_input_channels
aurora/transfer_function/base.py:225:        return self.tf_header.num_output_channels
aurora/transfer_function/base.py:297:                "output_channel": self.tf_header.output_channels,
aurora/transfer_function/base.py:298:                "input_channel": self.tf_header.input_channels,
aurora/transfer_function/base.py:302:        for out_ch in self.tf_header.output_channels:
aurora/transfer_function/base.py:303:            for inp_ch in self.tf_header.input_channels:
aurora/pipelines/process_mth5.py:329:    tf_collection = TransferFunctionCollection(header=tf_obj.tf_header, tf_dict=tf_dict)

rgreppy TransferFunctionHeader
yields:

aurora/sandbox/io_helpers/matlab_z_file_reader.py:16:from aurora.transfer_function.transfer_function_header import TransferFunctionHeader
aurora/sandbox/io_helpers/matlab_z_file_reader.py:35:    transfer_function_header = TransferFunctionHeader(
aurora/transfer_function/transfer_function_collection.py:4:1. TransferFunctionHeader
aurora/transfer_function/transfer_function_header.py:7:class TransferFunctionHeader(object):
aurora/transfer_function/base.py:43:    Header : transfer_function_header.TransferFunctionHeader() object.
aurora/pipelines/transfer_function_helpers.py:12:from aurora.transfer_function.transfer_function_header import TransferFunctionHeader
aurora/pipelines/transfer_function_helpers.py:64:    transfer_function_header = TransferFunctionHeader(
This was referenced Sep 28, 2021
Closed
Closed
@kkappler kkappler mentioned this issue Oct 6, 2021
Closed
@kkappler kkappler mentioned this issue Jun 20, 2022
Merged
20 tasks
@kkappler
Copy link
Collaborator Author

This is actually a somewhat handy convenience class, the issue is that it duplicates data that are available elsewhere, in particular, all the data in this object seems to come from the processing Config()

Thus, it would be best if this class were always generated by the processing config.
Doing this "Consolidation before deprecation" will allow deprecation to focusing only on that one generator.

The TransferFunctionHeader attrs that are called are the following:

  • processing_scheme (comes from tfk.config.decimations[0].estimator.engine)
  • _local_station: (comes from tfk.config.stations.local)
  • _reference_station: (comes from tfk.config.stations.remote)
  • input_channels: (comes from tfk.config.decimations[0].input_channels)
  • output_channels: (comes from tfk.config.decimations[0].output_channels)
  • reference_channels: (comes from tfk.config.decimations[0].reference_channels)

kkappler added a commit that referenced this issue Feb 11, 2023
@kkappler
Consolidate TransferFunctionHeader
4288536 
TransferFunctionHeader is now generated only by the Processing class.
This allowed the removal of a helper function that made the header
from processing.  Also added decimation level as an attr to the header.

[Issue(s): #109]
@kkappler
Copy link
Collaborator Author

Transfer Function Header is now consolidated into something that is generated as a method of Processing class.

The only change to still consider is making this a dict rather than a class - changing it to an mt_metadata object. However, that task will be more easily done while migrating the Processing class into mt_metadata which is o be addressed during the FC implementation.

@kkappler kkappler mentioned this issue Mar 31, 2023
Merged
@kkappler
Copy link
Collaborator Author

kkappler commented May 20, 2023 •
edited
Loading

If aurora's transfer_function class always has a processing_scheme instantiated when header is called, we can access header that way, via a method inside the transfer_function base class.
This will eliminate tf_header as an input argument, and simplify its usage.

To make this all work on the new fc branches, which should import the processing scheme from mt_metadata, and to avoid circular imports (from aurora) we will need to make TransferFunctionHeader a class that lives in mt_metadata.

A good candidate for this would be a ListDict (or python-box)

kkappler added a commit that referenced this issue May 20, 2023
@kkappler
Towards removing tf_header
e35a413 
- eliminate standalone calls to initialze tf_header
- make transfer function base class init in terms of (decimation level, processing scheme) rather than (tf_header, processing scheme)
- add tf_header as @Property of transfer function class
- make TransferFunctionHeader an mt_metadata ListDict rather than a standalone class

[Issue(s): #109]
kkappler added a commit that referenced this issue May 20, 2023
@kkappler
comment out failing test unrelated to issue #109
4f2cd38
kkappler added a commit that referenced this issue May 20, 2023
@kkappler
Fix matlab zfile reader test
91a8096 
- also commented out one test in fix_issue_139 so that all tests
will pass for PR

[Issue(s): #109]
kkappler added a commit that referenced this issue May 20, 2023
@kkappler
Deprecate TransferFunctionHeader
7fd7c91 
replace with method emtf_tf_header in processing class

[Issue(s): #109]
@kkappler
Copy link
Collaborator Author

This class has been completely deprecated and replaced with a ListDict that is returned by the emtf_tf_header method in processing.py

@kkappler kkappler mentioned this issue May 20, 2023
Merged
@kkappler
Copy link
Collaborator Author

This class has been completely deprecated and replaced with a ListDict that is returned by the emtf_tf_header method in processing.py

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kkappler