In [5]:

    

%load_ext autoreload
%autoreload 2
%matplotlib inline

import warnings
warnings.filterwarnings('ignore')

    
from misc_utils import load_raw_responses_with_traces,load_random_trace_sample
from session_and_tree_utils import *  


from response_distributions_util import plot_metric, get_single_motivation, plot_metric_breakdowns
import pandas as pd
import json
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

import scipy


    

    




The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

In [2]:

    

df = load_raw_responses_with_traces()
df = get_sessions_and_trees(df)
df['single motivation'] = df['motivation'].apply(get_single_motivation)


    

In [ ]:

    

df = df.apply(get_click_session, axis=1)
df['click_forest'] = df['click_session'].apply(getMinimumSpanningForest)
df = df.apply(get_click_tree, axis=1)


    

In [3]:

    

metric = 'num_pageviews'
df[metric] = df['click_session'].apply(num_pageviews)
plot_metric_breakdowns(df, metric)


    

    













    













    













    













    













    













    













    













    













    






<matplotlib.figure.Figure at 0x144aa0eb8>

In [4]:

    

metric = 'session_length'
df[metric] = df['click_session'].apply(get_session_length)
plot_metric_breakdowns(df, metric)


    

    













    













    













    













    













    













    













    













    













    






<matplotlib.figure.Figure at 0x13ba4eba8>

In [5]:

    

metric = 'external_searches'
df[metric] = df['click_session'].apply(external_searches)

plot_metric_breakdowns(df, metric)


    

    













    













    













    













    













    













    













    













    













    






<matplotlib.figure.Figure at 0x1445bca58>

In [6]:

    

from ml_utils import get_tree_metrics


    

In [7]:

    

tree_metrics = ['size', 'num_leafs', 'mean_time_diff' , 'degree_max' , 'depth_max']

for m in tree_metrics:
    df[m] = df['click_tree'].apply(lambda x: get_tree_metrics(x)[m])
    plot_metric_breakdowns(df, m)


    

    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    













    






<matplotlib.figure.Figure at 0x13165b978>

In [ ]: