In [ ]:

    

import os
import pandas as pd
import re
import subprocess
import sys
import matplotlib as mpl
mpl.use('Agg') 
import matplotlib.pyplot as plt
import seaborn as sns

%matplotlib inline


    

In [ ]:

    

sys.path.append('../code/')

from database_comparisons import Database, DatabaseComparison


    

In [ ]:

    

# for Waffle, until the default Java is set to 8:
#java="/usr/lib/jvm/java-8-oracle/jre/bin/java"
# for badger, set java = 'java'
java='java'


    

In [ ]:

    

! pwd


    

In [ ]:

    

! ls -l ../*.jar


    

In [ ]:

    

dbc = DatabaseComparison(desc_string='binary')
dbc.make_db(cutoff=0.04)


    

In [ ]:

    

dbc.summary


    

In [ ]:

    

dbc.make_db(cutoff=0.04)


    

In [ ]:

    

dbc.make_dbs([0.3, 0.5])


    

In [ ]:

    

dbc.summary


    

In [ ]:

    

dbc.databases


    

In [ ]:

    

p = dbc.plot_db_construction_time_vs_n_nodes()


    

In [ ]:

    

p = dbc.plot_db_construction_time_vs_cutoff()


    

In [ ]:

    

p = dbc.plot_cc_vs_cutoff()


    

In [ ]:

    

assert False


    

In [ ]:

    

def plot(df, x_col, y_col, filename=None):
    x = df[x_col]
    y = df[y_col]
    
    fig, ax = plt.subplots(1, 1, figsize=(4, 3))
    plt.plot(x, y, linestyle='--', marker='o', color='#756bb1')
    ax.set_ylim(bottom=0)
    
    #plt.legend(loc='best')
    ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))
    plt.xlabel(x_col)
    plt.ylabel(y_col)
    plt.title('Number of non-singleton connected components')
    plt.tight_layout()
    if filename:
        fig.savefig(filename)
    return fig


    

In [ ]:

    

connected_components = plot(results, 'cutoff', 'connected components')


    

In [ ]:

    

assert False


    

In [ ]:

    

os.path.dirname('../data_mining_Neo4j_v2_3_2/databases/db_binary_0.33/stderr_build.txt')


    

In [ ]:

    

! ls ../data_mining_Neo4j_v2_3_2/databases/db_binary_0.33


    

In [ ]:

    

db = Database(cutoff = 0.33, desc_string='binary')


    

In [ ]:

    

db.summary_df()


    

In [ ]:

    

load_existing_db(db)


    

In [ ]:

    

# try loading an existing db. 
db2 = Database(cutoff = 0.33, desc_string='binary')
db2.summary_df()


    

In [ ]:

    

load_existing_db(db)


    

In [ ]: