The bptools from pennmem

Errors when jacksheet numbers are over 256

Example traceback:

Traceback (most recent call last):
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/bin/ramulator-conf", line 11, in <module>
    load_entry_point('ramutils==2.1.10', 'console_scripts', 'ramulator-conf')()
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/ramutils/cli/expconf.py", line 196, in create_expconf
    use_classifier_excluded_leads=args.use_classifier_excluded_leads)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/ramutils/pipelines/ramulator_config.py", line 112, in make_ramulator_config
    use_common_reference).compute()
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/base.py", line 99, in compute
    (result,) = compute(self, traverse=False, **kwargs)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/base.py", line 206, in compute
    results = get(dsk, keys, **kwargs)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/threaded.py", line 75, in get
    pack_exception=pack_exception, **kwargs)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/local.py", line 521, in get_async
    raise_exception(exc, tb)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/compatibility.py", line 60, in reraise
    raise exc
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/local.py", line 290, in execute_task
    result = _execute_task(task, data)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/dask/local.py", line 271, in _execute_task
    return func(*args2)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/ramutils/tasks/_wrapper.py", line 22, in wrapper
    return func(*args, **kwargs)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/ramutils/tasks/odin.py", line 108, in generate_electrode_config
    ec.to_csv(csv_path)
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/bptools-1.3.0-py3.6.egg/bptools/odin/config.py", line 641, in to_csv
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/bptools-1.3.0-py3.6.egg/bptools/odin/config.py", line 595, in _export
  File "/home1/RAM_clinical/miniconda3/envs/ramutils/lib/python3.6/site-packages/bptools-1.3.0-py3.6.egg/bptools/odin/config.py", line 34, in _num_to_bank_label
KeyError: 4

Last few lines of jacksheet.txt:

220 RY14
221 RY15
222 RY16
223 ECG1
224 ECG2
258 DC09
259 DC10

area file parsing logic fails with extra whitespace

bptools/bptools/odin/config.py

Line 336 in 6203950

lines = [line.split() for line in f.read().split('\n') if len(line)]

We shouldn't check if len(line). Specifically this can be an issue if the only thing on a line is whitespace.

SeriesTransformation.monopolar_possible should be a property

standardize_label breaks for labels starting with numbers

This particularly affects Emory subjects (R1337E below):

Traceback (most recent call last):
  File "C:\Users\OdinUser\src\SYS3\rampy\ramulator\ramulator.py", line 1011, in load_experiment_config
    self.system_init()
  File "C:\Users\OdinUser\src\SYS3\rampy\ramulator\ramulator.py", line 1019, in system_init
    ret = self.init_experiment_runner()
  File "C:\Users\OdinUser\src\SYS3\rampy\ramulator\ramulator.py", line 400, in init_experiment_runner
    self.populate_electrode_table()
  File "C:\Users\OdinUser\src\SYS3\rampy\ramulator\ramulator.py", line 379, in populate_electrode_table
    self.configure_electrode_table(electrode_filename=electrode_filename, montage_filename=montage_filename)
  File "C:\Users\OdinUser\src\SYS3\rampy\ramulator\ramulator.py", line 798, in configure_electrode_table
    st = SeriesTransformation.create(electrode_filename, montage_file=montage_filename)
  File "c:\users\odinuser\src\bptools\bptools\transform.py", line 50, in create
    elec_conf = ElectrodeConfig(electrode_config_file)
  File "c:\users\odinuser\src\bptools\bptools\odin\config.py", line 248, in __init__
    self.read_config_file(filename)
  File "c:\users\odinuser\src\bptools\bptools\odin\config.py", line 417, in read_config_file
    contact.label = standardize_label(contact.label)
  File "c:\users\odinuser\src\bptools\bptools\util.py", line 43, in standardize_label
    label = ''.join(label.split())
AttributeError: 'long' object has no attribute 'split'

Standardizing labels still breaks Ramulator

Traceback (most recent call last):
  File "C:/Users/OdinUser/src/SYS3\rampy\ramulator\ramulator.py", line 1011, in load_experiment_config
    self.system_init()
  File "C:/Users/OdinUser/src/SYS3\rampy\ramulator\ramulator.py", line 1019, in system_init
    ret = self.init_experiment_runner()
  File "C:/Users/OdinUser/src/SYS3\rampy\ramulator\ramulator.py", line 400, in init_experiment_runner
    self.populate_electrode_table()
  File "C:/Users/OdinUser/src/SYS3\rampy\ramulator\ramulator.py", line 379, in populate_electrode_table
    self.configure_electrode_table(electrode_filename=electrode_filename, montage_filename=montage_filename)
  File "C:/Users/OdinUser/src/SYS3\rampy\ramulator\ramulator.py", line 798, in configure_electrode_table
    st = SeriesTransformation.create(electrode_filename, montage_file=montage_filename)
  File "c:\users\odinuser\src\bptools\bptools\transform.py", line 50, in create
    elec_conf = ElectrodeConfig(electrode_config_file)
  File "c:\users\odinuser\src\bptools\bptools\odin\config.py", line 256, in __init__
    self.read_config_file(filename)
  File "c:\users\odinuser\src\bptools\bptools\odin\config.py", line 425, in read_config_file
    contact.label = standardize_label(contact.label)
  File "c:\users\odinuser\src\bptools\bptools\util.py", line 59, in standardize_label
    if label[-2] == "0":
IndexError: string index out of range

Add ability to generate monopolar configurations

This should be done with a --monopolar flag.

Build conda packages

Implement an automated way to exclude non-brain channels from Jacksheet

Remove hard coded paths for SeriesTransformation tests

as_recarray methods improperly convert bytes to str

Example from:

bptools/bptools/odin/config.py

Lines 660 to 675 in 49a16ea

    
               def contacts_as_recarray(self): 
        
                   """Return the monopolar contacts as a recarry. 
        
                   This method exists for Ramulator compatibility. 
        
                   """ 
        
                   arr = np.recarray((len(self.contacts),), dtype=self._recarray_dtype) 
        
                   for i, contact in enumerate(self.contacts): 
        
                       arr[i]['jack_box_num'] = int(contact.port) 
        
                       arr[i]['contact_name'] = str(contact.label) 
        
                       # Not using surface area anymore because not used by Ramulator 
        
                       # arr[i]['surface_area'] = float(contact.area) 
        
                       arr[i]['description'] = str(contact.description) 
        
                   return arr

Line 670 turns a byte string into a str something like "b'LA1'" instead of "LA1".

Area file typos can lead to cryptic errors

bptools/bptools/odin/config.py

Lines 376 to 379 in e316f65

    
           areas = [ 
        
               area_map[area_map.label == row.electrode].area.values[0] 
        
               for _, row in js.iterrows() 
        
           ]

This can result in errors like:

IndexError: index 0 is out of bounds for axis 0 with size 0

when there are typos so that the labels in the area file don't match those in the jacksheet (e.g., LOFD instead of LOFCD). Workaround: catch the IndexError and provide a helpful hint about the problem before re-raising the exception.

Export Odin binary configuration file

Main changes:

~ as delimiter instead of ,
| as linebreak
integers encoded with <h
floats encoded with <f

pennmem / bptools Goto Github PK

bptools's People

Contributors

Watchers

bptools's Issues

Errors when jacksheet numbers are over 256

area file parsing logic fails with extra whitespace

SeriesTransformation.monopolar_possible should be a property

standardize_label breaks for labels starting with numbers

Standardizing labels still breaks Ramulator

Add ability to generate monopolar configurations

Build conda packages

Implement an automated way to exclude non-brain channels from Jacksheet

Remove hard coded paths for SeriesTransformation tests

as_recarray methods improperly convert bytes to str

Area file typos can lead to cryptic errors

Export Odin binary configuration file

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	def contacts_as_recarray(self):
	"""Return the monopolar contacts as a recarry.

	This method exists for Ramulator compatibility.

	"""
	arr = np.recarray((len(self.contacts),), dtype=self._recarray_dtype)

	for i, contact in enumerate(self.contacts):
	arr[i]['jack_box_num'] = int(contact.port)
	arr[i]['contact_name'] = str(contact.label)
	# Not using surface area anymore because not used by Ramulator
	# arr[i]['surface_area'] = float(contact.area)
	arr[i]['description'] = str(contact.description)

	return arr

	areas = [
	area_map[area_map.label == row.electrode].area.values[0]
	for _, row in js.iterrows()
	]