import pandas as pd
orig_data = pd.read_csv("ZDO2014evaluation.csv")
#orig_data.sort(column=['score'], ascending=False)
#print orig_data
gb = orig_data.groupby('team')
#print gb.max().sort(column=['score'], ascending=False)

gbsort = gb.max().sort(column=['score'], ascending=False)
gbsort['score']
#tt.tolist()

%pylab inline --no-import-all
import matplotlib.pyplot as plt

print np.power(gbsort['score'].astype(np.double),2)*36

## vizualizace progrese
te = np.arange(0,1.1,0.1)
y = np.power(te,2)*36
plt.figure(figsize=(2,1.5))
plt.plot(te,y)

#pd.set_option('display.max_columns', None)
print orig_data.tail(40)
#print dir(pd)

import logging
logger = logging.getLogger(__name__)

logging.basicConfig(level=logging.DEBUG)

logger.debug('muj vypis cislo ' + str(1))

%pylab inline --no-import-all
import sklearn
import sklearn.metrics
import matplotlib.pyplot as plt

df = pd.read_csv("ZDO2014classifs.csv")

cols = df.columns


print cols

# získáme seznam všech labelů značek, které jsou v testu
unlabels = np.unique(df['reference'].values)
# převedeme do čísel (indexů)
y_true = np.searchsorted(unlabels, df['reference'].values)
print unlabels

for col in cols:
    y_pred = np.searchsorted(unlabels, df[col].values)
    cmat = sklearn.metrics.confusion_matrix(y_true, y_pred)
    plt.matshow(cmat)
    plt.title(col)