update.py

# -*- coding: utf-8 -*-
""" Update Readme.md and cumulative_gans.jpg """
from __future__ import print_function
from __future__ import division
from wordcloud import WordCloud
from wordcloud import STOPWORDS

import numpy as np
import matplotlib.pyplot as plt
import sys
import datetime
import pandas as pd
import csv
import json
import os

def load_data():
    """ Load GANs data from the AllGAN.csv file """
    import csv
    import codecs

    with codecs.open('AllGAN-r2.tsv',"rbU", "utf-8") as fid:
        reader = csv.DictReader(fid, delimiter='\t')
        gans = [row for row in reader]
    return gans


def update_readme(gans):
    """ Update the Readme.md text file from a Jinja2 template """
    import jinja2 as j2

    gans.sort(key=lambda v: v['Title'].upper())
    j2_env = j2.Environment(loader=j2.FileSystemLoader('.'),
                            trim_blocks=True, lstrip_blocks=True)

    j2_env.globals['nowts'] = datetime.datetime.now()

    with open('README-one.md', 'w') as fid:
        print(j2_env.get_template('README.j2.md').render(gans=gans), file=fid)

def update_index(gans):
    """ Update the index.html text file from a Jinja2 template """
    import jinja2 as j2

    try:
        gans.sort(key=lambda v: ((int(v['Year']) if v['Year'].isdigit() else v['Year'])
        , (int(v['Month']) if v['Month'].isdigit() else v['Month'])), reverse=True)
    except:
        pass
    j2_env = j2.Environment(loader=j2.FileSystemLoader('.'),
                            trim_blocks=True, lstrip_blocks=True)

    j2_env.globals['nowts'] = datetime.datetime.now()

    with open('docs/index.html', 'w') as fid:
        print(j2_env.get_template('INDEX.j2.md').render(gans=gans), file=fid)


def update_figure(gans):
    """ Update the figure cumulative_gans.jpg """
    data = np.array([int(gan['Year']) + int(gan['Month']) / 12
                     for gan in gans])
    x_range = int(np.ceil(np.max(data) - np.min(data)) * 12) + 1
    y_range = int(np.ceil(data.size / 10)) * 10 + 1

    with plt.style.context("seaborn"):
        plt.hist(data, x_range, cumulative="True")
        plt.xticks(range(2014, 2018))
        plt.yticks(np.arange(0, y_range, 15))
        plt.title("Cumulative number of named GAN papers by month")
        plt.xlabel("Year")
        plt.ylabel("Total number of papers")
        plt.savefig('cumulative_gans.jpg')

def update_wordcloud_title():
    """ Update the figure wordcloud_title.jpg """

    data = pd.read_csv('AllGAN-r2.tsv',delimiter='\t', encoding='utf-8')

#    tmp_data = data['Title'].split(" ") for x in data

#    count_list = list([list(x) for x in data['Title'].value_counts().reset_index().values])

#    wordcloud = WordCloud(stopwords=STOPWORDS,relative_scaling = 0.2,
#                        max_words=2000, background_color='white').generate_from_frequencies(tmp_data)
    stopwords = set(STOPWORDS)
    #ganstop = ['Generative','Adversarial', 'Networks', 'Network', 'GAN', 'GANs', 'using', 'Learning', 'Training', 'Generation',
    #        'Neural', 'Net', 'Model', 'Nets', 'Deep', 'Based', 'Via', 'Conditional', 'Models', 'Examples']
    #stopwords.add(ganstop)

    stopwords.add('Generative')
    stopwords.add('Adversarial')
    stopwords.add('Networks')
    stopwords.add('Network')
    stopwords.add('GAN')
    stopwords.add('GANs')
    stopwords.add('using')
    stopwords.add('Learning')
    stopwords.add('Training')
    stopwords.add('Generation')
    stopwords.add('Neural')
    stopwords.add('Net')
    stopwords.add('Model')
    stopwords.add('Nets')
    stopwords.add('Deep')
    stopwords.add('Based')
    stopwords.add('Via')
    stopwords.add('Conditional')
    stopwords.add('Models')
    stopwords.add('Examples')

    wordcloud = WordCloud(stopwords=stopwords,relative_scaling = 0.2, random_state=3,
                    max_words=2000, background_color='white').generate(' '.join(data['Title']))

    plt.figure(figsize=(12,12))
    plt.imshow(wordcloud, interpolation="bilinear")
    plt.axis("off")
    #plt.show()
    #plt.savefig('wordcloud_title.png')
    wordcloud.to_file('wordcloud_title.png')
    wordcloud.to_file('docs/png/wordcloud_title.png')

def update_wordcloud_category():
    """ Update the figure wordcloud_category.jpg """

    data = pd.read_csv('AllGAN-r2.tsv',delimiter='\t', encoding='utf-8')

    wordcloud = WordCloud(stopwords=STOPWORDS,relative_scaling = 0.2, random_state=3,
                max_words=2000, background_color='white').generate(' '.join(data['Category']))

    plt.figure(figsize=(12,12))
    plt.imshow(wordcloud, interpolation="bilinear")
    plt.axis("off")
    #plt.show()
    #plt.savefig('wordcloud_title.png')
    wordcloud.to_file('wordcloud_category.png')
    wordcloud.to_file('docs/png/wordcloud_category.png')

def update_wordcloud_abbr():
    """ Update the figure wordcloud_category.jpg """

    data = pd.read_csv('AllGAN-r2.tsv',delimiter='\t', encoding='utf-8')

    wordcloud = WordCloud(stopwords=STOPWORDS,relative_scaling = 0.2, random_state=3,
                max_words=2000, background_color='white').generate(' '.join(data['Abbr.']))

    plt.figure(figsize=(12,12))
    plt.imshow(wordcloud, interpolation="bilinear")
    plt.axis("off")
    #plt.show()
    #plt.savefig('wordcloud_title.png')
    wordcloud.to_file('wordcloud_abbr.png')
    wordcloud.to_file('docs/png/wordcloud_abbr.png')

def update_csv2json():

#    COLUMNS = ('Mnum','Abbr.','Title','Year','Month','Citations','pdf','Arxiv','Official_Code','Tensorflow','PyTorch','KERAS',	'Stars','Web','No','SN','Medical','Category')
    with open('AllGan-r2.tsv', 'r') as f:
        reader = csv.DictReader(f, delimiter='\t')
        rows = list(reader)

    with open('docs/AllGan.json', 'w') as f:
        f.write(json.dumps(rows, sort_keys=False, separators=(',', ': '), ensure_ascii=False, indent=4))

if __name__ == '__main__':
    try:
        reload(sys)  # Python 2
        sys.setdefaultencoding('utf-8')
    except NameError:
        pass         # Python 3

    GANS = load_data()
    update_wordcloud_title()
    update_wordcloud_category()
    update_wordcloud_abbr()
    update_readme(GANS)
    update_index(GANS)
    update_csv2json()
#    update_figure(GANS)