In [1]:

    

import karps as ks
import karps.functions as f
from karps.display import show_phase


    

In [2]:

    

df = ks.dataframe([1.0, 2.0], name="my_input")
df


    

    
Out[2]:





/my_input@org.spark.DistributedLiteral:double

In [3]:

    




    

In [4]:

    

def harmonic_mean(col):
    count = f.as_double(f.count(col))
    inv_sum = f.inv(f.sum(f.inv(col)))
    return inv_sum * count


    

In [5]:

    

cached_df = f.autocache(df)
hmean = harmonic_mean(cached_df)
hmean


    

    
Out[5]:





/multiply_11!org.spark.LocalStructuredTransform:double

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [6]:

    

s = ks.session("demo1c")
comp = s.compute(hmean)


    

In [7]:

    

show_phase(comp, "initial")


    

In [8]:

    

show_phase(comp, "REMOVE_OBSERVABLE_BROADCASTS")


    

In [9]:

    

show_phase(comp, "MERGE_PREAGG_AGGREGATIONS")


    

In [10]:

    

show_phase(comp, "MERGE_AGGREGATIONS")


    

In [11]:

    

show_phase(comp, "final")


    

In [12]:

    

comp.values()


    

    
Out[12]:





(double, double_value: 1.3333333333333333
)

In [13]:

    

s.run(the_mean)


    

    




---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-13-d40191459ce7> in <module>()
----> 1 s.run(the_mean)

NameError: name 'the_mean' is not defined

In [ ]: