def graph_present():
data = load_rep_data()
df = pd.DataFrame(data)
sub = df[['name', 'present', 'yea', 'nay', 'absent']].sort_values(by='yea', ascending=False)
sns.barplot(x="yea", y="name", data=sub, label='small')
plt.show()
return
def graph_present_by_party():
data = load_rep_data()
df = pd.DataFrame(data)
sub = df[['name', 'present', 'party', 'yea', 'nay']]
group = sub.groupby(['party']).sum().sort_values(by='present', ascending=True)
group.plot(kind='barh', ax=ax)
plt.show()
return
def make_stat_array_votes():
data = load_rep_data()
votes = [x['votes'] for x in data]
names = [x['name'] + ' (' + x['party'] + ')' for x in data]
sizes = [x['present'] for x in data]
parties = [x['party'] for x in data]
result = np.array(votes)
return {'mat': votes, 'names': names, 'sizes': sizes, 'parties': parties}
def plot_solidarity_index():
data = load_rep_data()
mat = []
for d in data:
mat.append(d['votes'])
line = np.sum(mat, axis=0)
plt.matshow(line)
print line.shape
plt.show()
return
def show_votes():
data = load_rep_data()
mat = []
labels = []
for d in data:
mat.append(d['votes'][:100])
labels.append(d['name'])
mat = np.array(mat)
ax.matshow(mat, cmap=plt.cm.get_cmap('coolwarm'), aspect='auto')
ax.set_yticklabels(['']+labels)
plt.show()
return
def text_to_vectors():
data = load_rep_data()#[:40]
names = [x['name'] + ' (' + x['party'] + ')' for x in data]
quotes = [x['quotes'].lower() for x in data]
sizes = [x['present'] for x in data]
parties = [x['party'] for x in data]
vectorizer = TfidfVectorizer\
(min_df=25, stop_words=stop_words,
strip_accents='unicode', lowercase=True, ngram_range=(1, 2),
norm='l2', smooth_idf=True, sublinear_tf=False, use_idf=True,
analyzer='word'
)
print 'vectorizing ...'
X = vectorizer.fit_transform(quotes)
D = -(X * X.T).todense()
return {'mat': D, 'names': names, 'sizes': sizes, 'parties': parties}
