batchent.py

import entropy
import os
import csv
import target_ngrams
import imp
imp.reload(target_ngrams)


def make_info_graphs(directory='./data/BFs_logan/data', nrange=[2,7], shuffle_mode=False, min_count=1, backwards=False, ExpectedProbability=False):
    LOL = []
    mode = 'w'
    for filename in os.listdir(directory):
        info_profile = entropy.avg_ent(
            directory + '/' + filename, nrange, shuffle_mode, min_count, backwards = backwards, ExpectedProbability = ExpectedProbability)
        row = [filename]
        for value in info_profile.values():
            row.append(value)
        with open("./output/batchent.csv", mode) as output_file:
            writer = csv.writer(output_file)
            writer.writerow(row)
        mode = 'a'


def batch_syl_info_and_feats(
        directory='./data/BFs_logan/data',
        n=2,
        spectral_fp='./data/spectral.csv',
        prefix='fathers_and_sons_from_logan - ',
        backwards=False, ExpectedProbability=False):
    LOL = []
    mode = 'w'
    for filename in os.listdir(directory):
        fn_listchar = list(filename[:-4])
        pf_listchar = list(prefix)
        for char in pf_listchar:
            fn_listchar.remove(char)
        bird_ID = ''
        for char in fn_listchar:
            bird_ID += char
        tn_result = target_ngrams.info_and_feats(
            directory + '/' + filename, n, spectral_fp, bird_ID, backwards = backwards, ExpectedProbability = ExpectedProbability)
        for row in tn_result:
            row.insert(0, bird_ID)
            with open("./output/batch_syl_info_and_feats.csv", mode) as output_file:
                writer = csv.writer(output_file)
                writer.writerow(row)
            mode = 'a'
        for item in tn_result:
            if item not in LOL:
                LOL.append(item)
    return LOL


# have to get rid of no_data rows, before using this for spectral analysis
def nest_comparison(
        bird_IDs,
        directory,
        n,
        spectral_fp,
        index,
        prefix='fathers_and_sons_from_logan - ',
        previous_result='None',
        nest_name=''):
    if isinstance(previous_result, list):
        pass
    else:
        previous_result = batch_syl_info_and_feats(
            directory, n, spectral_fp, prefix)
    bird_syls = []
    for row in previous_result:
        bird_syls.append(row[:2])
    common_syllables = []
    bird1 = bird_IDs[0]
    bird1_syllables = []
    for syl in bird_syls:
        if syl[0] == bird1 and syl not in bird1_syllables:
            bird1_syllables.append(syl[1])
    current_list = bird1_syllables
    for bird in bird_IDs[1:]:
        print(bird)
        current_bird_syllables = []
        for syl in bird_syls:
            if syl[0] == bird and syl not in current_bird_syllables:
                current_bird_syllables.append(syl[1])
        current_list = list(
            set(current_list).intersection(current_bird_syllables))
    current_list.sort()
    header = bird_IDs
    header.insert(0, 'Syllable')
    data = [header]
    for syllable in current_list:
        minilist = [syllable]
        birds_seen = []
        for row in previous_result:
            if row[0] in bird_IDs and row[1] == syllable and row[0] not in birds_seen:
                minilist.append(row[index])
                birds_seen.append(row[0])
        data.append(minilist)
    mode = 'w'
    for row in data:
        with open("./data/nest_comparison/%s_nest_comparison.csv" % nest_name, mode) as output_file:
            writer = csv.writer(output_file)
            writer.writerow(row)
        mode = 'a'
    return data