Pandas utility scripts

clean.py

#check if na values exist at all
df.isnull().any().any()
#if above returns true for any series
df.isnull().any()

#check for NA values
data.isnull().sum()

#replace NA values
# for real values
data['series_to_replace'] = data['series_to_replace'].fillna(data.mean()['series_to_replace'])

# convert string to int
mapping = {string:int}
data.replace({'column_name':mapping})

#find number of unique items in a series
data['series_name'].nunique()
#find uniquevalues
pd.unique(data['column_name'])
#this http://stackoverflow.com/questions/17995328/changing-values-in-pandas-dataframe-doenst-work