def plotAvgHist(column, field, bins=40):
global subplot
subplot +=1
plt.subplot(subplot)
plt.title( '%s: \nmean=%s std=%s ' % (field, round(mean(column), 3), round(std(column), 3)), fontsize=11 )
plt.axhspan(0, 2000, color='none')
plt.hist( column, bins=bins, range=(0,1))
print '%s: mean=%s std=%s items_under_0.3=%s' % (field, round(mean(column), 3), round(std(column), 3), sum([1 for c in column if c<0.3]) )
JCR = JamendoCsvReader('stats_reviewAVGs.csv')
subplot = 220
columns = JCR.getColumns(['reviews_avgnote', 'avg_agreed_note', 'weighted_avg_note', 'weighted_avg_agreed_note', 'weighted_avg_agreed_note2', 'reviews_all'], \
filterfunc=filterfieldsunder(['reviews_all'],4))
plt.figure(figsize=(12,8))
for key in ['avg_agreed_note', 'weighted_avg_note', 'weighted_avg_agreed_note', 'weighted_avg_agreed_note2']:
plotAvgHist(columns[key], key)
plt.figure()
plt.title( '%s: \nmean=%s std=%s ' % ('reviews_avgnote', round(mean(columns['reviews_avgnote']), 3), \
from sys import path
path.append('../')
from core.JamendoCsvReader import JamendoCsvReader
import matplotlib.pyplot as plt
from numpy import array
JCR = JamendoCsvReader('stats_album_total.csv')
#AGGREGATE COLUMNS TO MAKE LINEAR OPERATION LIKE SUM
FBcols = JCR.getColumns(['fb_likesharecomment', 'starred'])
total_likes = FBcols['fb_likesharecomment'] + FBcols['starred']
plt.figure()
plt.plot(total_likes, 'b-', linewidth=1)
plt.title('fb total (sum of likes,comments and shares) + starred')
#COMPARE CURVE WITHOUT USING compareJoinedColumnsPlotting (you have more freedom to directly set any parameters...)
JC = JCR.getColumns(['reviews_avgnote', 'weighted_avg_agreed_note'], sortkey='reviews_avgnote')
avgweightednote = JC['weighted_avg_agreed_note']
avgnote = JC['reviews_avgnote']
plt.figure()
from sys import path
path.append('../')
from core.JamendoCsvReader import JamendoCsvReader
import matplotlib.pyplot as plt
import numpy as np
JCR = JamendoCsvReader('stats_album_total.csv')
plt.figure(figsize=(12,8))
plt.subplot(211)
avgnote = JCR.getColumnArray('reviews_avgnote')
avgnote_no0 = JCR.getColumnArray('reviews_avgnote', filterfunc=lambda x:True if x>0.0 else False)
avgnote_0 = len(JCR.getColumnArray('reviews_avgnote', filterfunc=lambda x:True if x==0.0 else False))
plt.hist( avgnote, bins=10, range=(0,1))
plt.hist( avgnote_no0, bins=10, range=(0,1))
plt.title('Histogram with the distribution of review_avgnotes, spread on ten bins of length 0.5. \n\
The green one exclude items with review avg=0, so the ones that \n probably have no votes (%s)' % avgnote_0, fontsize=12)
avgnote = JCR.getColumnArray('weighted_avg_agreed_note')
avgnote_no0 = JCR.getColumnArray('weighted_avg_agreed_note', filterfunc=lambda x:True if x>0.0 else False)
avgnote_0 = len(JCR.getColumnArray('weighted_avg_agreed_note', filterfunc=lambda x:True if x==0.0 else False))
