SF_AC_Name_4 (3).py

import pandas as pd
import numpy as np
from fuzzywuzzy import process
import multiprocessing as mp

def multi_run_wrapper(args):
    return process_df(*args)


def find_app_name(name, all_names):

    #### set score_cutoff =88 for Magnus ###
    try:
        app_name = process.extractBests(name, choices=all_names, score_cutoff=88, limit=5)
        result = [name[0] for name in app_name if 88 <= name[1] < 100]

        if len(result) == 0:
            #                 print("Skipping {}, found no similiar match".format(name))
            return np.nan

        final_result = "".join(str(item) for item in result)
        return final_result
        # return result
    except Exception as e:
        print(e)
        pass
    return np.nan


def process_df(df, all_names):
    df[0]["Suggested Name"] = df[0].apply(lambda x: find_app_name(x["Account Name"], all_names), axis=1)
    return df


if __name__ == '__main__':
    infileName = 'SF Account Names data.csv'
    outfileName = "SF Account Names data - out.xlsx"

    sfData = pd.read_csv(infileName, header=0)
    Name = sfData["Account Name"]
    sfData["Suggested Name"] = np.nan

    cpu_count = mp.cpu_count()
    p = mp.Pool(cpu_count)
    pool_results = p.map(multi_run_wrapper, [(np.array_split(sfData, cpu_count), Name)])
    p.close()
    p.join()

    results = pd.concat(pool_results[0])
    sf_output = results[~pd.isnull(results['Suggested Name'])]
    # print(sf_output.head(10))

    writer = pd.ExcelWriter(outfileName)
    sf_output.to_excel(writer, 'SF Account Name')
    writer.save()