In [ ]:

    

import ck.kernel as ck
import pandas as pd
import numpy as np
import matplotlib as matplotlib
import matplotlib.pyplot as plt
import json
import os


    

In [ ]:

    

print "Collective Knowledge: v%s" % ck.version({})['version_str']
print "pandas: v%s" % pd.__version__
print "NumPy: v%s" % np.version.version
print "Matplotlib: v%s" % matplotlib.__version__
print "JSON: v%s" % json.__version__


    

In [ ]:

    

dataset = 'SGEMM_NT'
data_uoa =  dataset + '-explore-f-n'
module_uoa = 'experiment'


    

In [ ]:

    

r=ck.access({'action':'list_points', 'module_uoa':module_uoa, 'data_uoa':data_uoa})
if r['return']>0:
  print ("Error: %s" % r['error'])
  exit(1)


    

In [ ]:

    

data_list  = []
index_list = []

for point in r['points']:
    with open(os.path.join(r['path'], 'ckp-%s.flat.json' % point)) as point_file:
        point_data = json.load(point_file)
    # Data.    
    Gflops_per_s = point_data.get("##characteristics#run#run_time_state#EXECUTION#Gflops/s#all")
    data_list.append(Gflops_per_s)
    # Index.
    cl_file = point_data.get("##characteristics#run#run_time_state#METADATA#file#all_unique")[0]
    lws_j = point_data.get("##characteristics#run#run_time_state#EXECUTION#lws_j#all_unique")[0]
    lws_i = point_data.get("##characteristics#run#run_time_state#EXECUTION#lws_i#all_unique")[0]
    local_work_size = ('(%s, %s)' % (lws_j, lws_i))
    matrix_order = point_data.get("##characteristics#run#run_time_state#CMD_LINE_ARGS#matrix_order#all_unique")[0]
    index_list.append((cl_file, local_work_size, matrix_order))


    

In [ ]:

    

mi = pd.MultiIndex.from_tuples(names=['OpenCL file', 'Local work size', 'Matrix order'], tuples=index_list)
df = pd.DataFrame(data=data_list, index=mi).sortlevel(level='OpenCL file')


    

In [ ]:

    

df


    

In [ ]:

    

%matplotlib inline


    

In [ ]:

    

mean = df.mean(axis=1)
std = df.std(axis=1)
ymax = np.int64(mean.max() + std.max())


    

In [ ]:

    

mean.unstack(level='Local work size').unstack(level='OpenCL file')


    

In [ ]:

    

std.unstack(level='Local work size').unstack(level='OpenCL file')


    

In [ ]:

    

# Issue with Matplotlib 1.5?
# e.g. see http://stackoverflow.com/questions/34128232/keyerror-when-trying-to-plot-or-histogram-pandas-data-in-matplotlib

# plot = mean.unstack(level='Local work size').unstack(level='OpenCL file') \
#     .plot(yerr=std.unstack(level='Local work size').unstack(level='OpenCL file'),
#         title='Gflops/s vs Matrix order',
#         kind='bar', figsize=(12,8), colormap=matplotlib.cm.autumn,
#         ylim=(0, ymax), yticks=range(ymax))


    

In [ ]:

    

plot = mean.unstack(level='Local work size').unstack(level='OpenCL file') \
    .plot(
        title='Gflops/s vs Matrix order',
        kind='bar', figsize=(12,8), colormap=matplotlib.cm.autumn,
        ylim=(0, ymax), yticks=range(ymax))
plot


    

In [ ]:

    

# A bug in pandas 0.16.1 prevents us from using MultiIndex when dumping to LaTeX.
mi_tex = pd.MultiIndex.from_tuples(tuples=index_list)
df_tex = pd.DataFrame(data=data_list, index=mi_tex).sortlevel(level=0)
# Add mean and std columns.
df_tex['mean'] = mean
df_tex['std'] = std

with open('%s_tmp.tex' % data_uoa, 'w') as tex_file:
    tex_file.write(df_tex.to_latex())


    

In [ ]:

    

plot.get_figure().savefig('%s_tmp.pdf' % data_uoa)
plot.get_figure().savefig('%s_tmp.png' % data_uoa)