Ticket/2454/rename/vbn/fields

[danielsf]

This PR renames some fields in the NWB and metadata tables to bring the two datastreams into alignment.

It also adds the "strip-substructure" functionality to the "structure_acronym" column in the metadata tables.

[aamster]

Looks good, with a few minor comments

[aamster]

It looks like the outer if can be safely removed since you are just checking if it is nan here

[danielsf]

Actually, numpy doesn't like running np.isnan on non-numbers

>>> import numpy as np
>>> np.isnan('abcd')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: ufunc 'isnan' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''
>>> 

[aamster]

It looks like this part can recursively call strip_substructure_acronym and add the results to a list if you feel comfortable writing that

[danielsf]

Good catch; done.

[aamster]

I'm not sure why this function is needed. Can you not just do

df[col_name] = df[col_name].apply(strip_substructure_acronym)

[danielsf]

I can; I just don't know that much pandas. I've made the change.

[aamster]

Note that apply is not an inplace operation. ie

units_table['structure_acronym'].apply(strip_substructure_acronym)

won't have any effect. It needs to be assigned.

[danielsf]

this is why I hate pandas ;)

[danielsf]

Just running apply results in a series....

>>> import pandas as pd
>>> data = [{'a': 1, 'b': 2}, {'a': 2, 'b': 3}]
>>> df = pd.DataFrame(data=data)
>>> def fn(x):
...     return x**2
... 
>>> 
>>> new_df = df['a'].apply(fn)
>>> new_df
0    1
1    4
Name: a, dtype: int64

I'm just going to go back to my "implement a function to munge the structure acronym column." It was tested.

I really do not like pandas

[aamster]

You need to assign the output to a column, not to the dataframe, i.e

units_table['structure_acronym'] = units_table['structure_acronym'].apply(strip_substructure_acronym)