hipom_data_mapping/data_preprocess/abbreviations/abbreviations_replacer.py

79 lines
2.8 KiB
Python

"""
Author: Daniel Kim
Modified by: Richard Wong
"""
# %%
import re
import pandas as pd
from replacement_dict import desc_replacement_dict, unit_replacement_dict
# %%
def count_abbreviation_occurrences(tag_descriptions, abbreviation):
"""Count the number of occurrences of the abbreviation in the list of machine descriptions."""
pattern = re.compile(abbreviation)
count = sum(len(pattern.findall(description)) for description in tag_descriptions)
return count
def replace_abbreviations(tag_descriptions, abbreviations):
"""Replace the abbreviations according to the key-pair value provided."""
replaced_descriptions = []
for description in tag_descriptions:
for abbreviation, replacement in abbreviations.items():
description = re.sub(abbreviation, replacement, description)
replaced_descriptions.append(description)
return replaced_descriptions
def cleanup_spaces(tag_descriptions):
# Replace all whitespace with a single space
replaced_descriptions = []
for description in tag_descriptions:
description_clean = re.sub(r'\s+', ' ', description)
replaced_descriptions.append(description_clean)
return replaced_descriptions
# remove all dots
def cleanup_dots(tag_descriptions):
replaced_descriptions = []
for description in tag_descriptions:
description_clean = re.sub(r'\.', '', description)
replaced_descriptions.append(description_clean)
return replaced_descriptions
# %%
file_path = '../../data_import/exports/raw_data.csv' # Adjust this path to your actual file location
df = pd.read_csv(file_path)
# %%
# Replace abbreviations
print("running substitution for descriptions")
df['tag_description']= df['tag_description'].fillna("NOVALUE")
# Replace whitespace-only entries with "NOVALUE"
# note that "N/A" can be read as nan
# replace whitespace only values as NOVALUE
df['tag_description'] = df['tag_description'].replace(r'^\s*$', 'NOVALUE', regex=True)
tag_descriptions = df['tag_description']
replaced_descriptions = replace_abbreviations(tag_descriptions, desc_replacement_dict)
replaced_descriptions = cleanup_spaces(replaced_descriptions)
replaced_descriptions = cleanup_dots(replaced_descriptions)
df["tag_description"] = replaced_descriptions
# print("Descriptions after replacement:", replaced_descriptions)
# strip trailing whitespace
df['tag_description'] = df['tag_description'].str.rstrip()
df['tag_description'] = df['tag_description'].str.upper()
# %%
print("running substitutions for units")
df['unit'] = df['unit'].fillna("NOVALUE")
df['unit'] = df['unit'].replace(r'^\s*$', 'NOVALUE', regex=True)
unit_list = df['unit']
new_unit = replace_abbreviations(unit_list, unit_replacement_dict)
new_unit = cleanup_spaces(new_unit)
df['unit'] = new_unit
# save
df.to_csv("../exports/preprocessed_data.csv", index=False)
print("file saved")