ax.set_title('Distance between Word Instances by Book and Author')
ax.set_xticks(index + bar_width / 2)
plt.xticks(fontsize=8, rotation=90)
ax.set_xticklabels(book)
ax.legend()
fig.tight_layout()
#plt.show()
filepng = "visualizations/" + title + ".png"
filepdf = "visualizations/" + title + ".pdf"
plt.savefig(filepng, bbox_inches='tight')
plt.savefig(filepdf, bbox_inches='tight')
if __name__ == '__main__':
corpus = Corpus('sample_novels')
novels = corpus._load_novels()
num = 0
#while num <10:
medians_he = []
medians_she = []
books = []
for novel in novels[num * 10:num * 10 + 9]:
result_he = instance_dist(novel, "he")
result_she = instance_dist(novel, "she")
try:
medians_he.append(median(result_he))
except:
medians_he.append(0)
>>> find_female_adj(scarlett)
{'beautiful': 3, 'sad': 1}
:param:novel
:return: dictionary of adjectives that appear around female pronouns and the number of occurences
"""
return find_gender_adj(novel, True)
if __name__ == '__main__':
test_function()
print("loading corpus")
corpus = Corpus('sample_novels')
print("loading novel")
novel = corpus._load_novels()[15]
print(novel.author, novel.title, novel.word_count)
print("running function")
result = find_male_adj(novel)
output = []
for key in result.keys():
output.append((result[key], key))
print(sorted(output, reverse=True))
def process_medians(helst, shelst, authlst):
"""
>>> medians_he = [12, 130, 0, 12, 314, 18, 15, 12, 123]
>>> medians_she = [123, 52, 12, 345, 0, 13, 214, 12, 23]
>>> books = ["a", "b", "c", "d", "e", "f", "g", "h", "i"]
>>> process_medians(helst=medians_he, shelst=medians_she, authlst=books)