def print_other(data):
stopwords = ["replication", "topmine", "nltk"]
headers = ["ToPMine \#topics", "ToPMine $\\alpha$", "Stopwords"]
latex_table.print_latex_table(
data=[data],
header=headers,
caption="Parameters used that were not specified by the authors",
color="blue")
def print_z_score_calculation_table(calc_info, n, reverse=True):
'''
prints top n event candidates by z-score
'''
top_list = sorted(calc_info, key=operator.itemgetter(1), reverse=True)[:n]
#print latex table for the event candidates
header = [
"event phrases", "z_score", "cluster_weight", "$\\mu_{T}$",
"$\\sigma_{T}$"
]
latex_table.print_latex_table(top_list, header=header, alignment="lrrrr")
def print_specified(data):
# headers = ["ToPMine minsupport", "ToPMine n-gram", "FPgrowth minsupport", "$\\theta$", "$\omega$", "$\\alpha$", "$\\beta$", "$\chi$", "$\\tau$"]
headers = [
"ToPMine minsupport", "ToPMine n-gram", "FPgrowth minsupport",
"$\\theta$", "$\omega$", "$\\alpha$", "$\\beta$", "$\chi$", "$\\tau$"
]
default = ["40", "5", "8", "3", "0.1", "10", ".05", ".5", "24hr"]
# experiment = ["15", "5", "1", "2", "0.1", "10", ".05", ".5", "24hr"]
latex_table.print_latex_table(data=[default, data],
header=headers,
caption="Paper specified parameters",
color="cyan")
def graph_n_best(timestep_measure_event, n, headers):
data = []
# just so I save all the results:
# n = len(toplist)
if n > len(toplist):
n = len(toplist)
for t, e, v in toplist[:n]:
data.append(
[timestep_to_datetime[event],
str(round(val, 2)),
str(event)])
if len(data) > 0:
latex_table.print_latex_table(data, header=headers, alignment="llr")
return
def print_z_score_table(event_candidates, n, reverse=True):
'''
prints top n event candidates by z-score
'''
top_list = sorted(event_candidates,
key=operator.itemgetter(0),
reverse=reverse)[:n]
# #find what the max number of phrases is for this list
# pkl_max = max([len(l[1]) for l in sorted(top_list)])
# print(pkl_max)
#print latex table for the event candidates
top_n_list = []
for value, phrase_keys in top_list:
l = list(phrase_keys)
top_n_list.append([value] + [', '.join(l)])
header = ["z_score", "event phrases"]
latex_table.print_latex_table(top_n_list, header=header, alignment="rl")
def full_analysis(clusters, timestep_to_datetime, n):
'''
clusters[phrase] = {"alpha":num_peaks[phrase],"beta":stds[phrase],"chi":intensities[phrase]}
'''
if n > len(clusters):
n = len(clusters)
# #graph by smallest number of peaks
# alphas = {event:clusters[event]["alpha"] for event in clusters}
# sorted_alphas = sorted(alphas.items(), key=operator.itemgetter(1))
# timestep_measure_event = []
# for e,v in sorted_alphas[:n]:
# timestep_measure_event.append([timestep_to_datetime[clusters[e]["timestep"]], "{}".format(v), "{:.3f}".format(clusters[e]["beta"]), "{:.3f}".format(clusters[e]["chi"]), e])
# latex_table.print_latex_table(data=timestep_measure_event, header=["timestep", "$\\alpha$","$\\beta$", "$\chi$", "event"], alignment="llllp{12.0cm}", caption="$\\alpha$, the number of peak heights", color="green")
# betas = {event:clusters[event]["beta"] for event in clusters}
# sorted_betas = sorted(betas.items(), key=operator.itemgetter(1), reverse=True)
# timestep_measure_event = []
# for e,v in sorted_betas[:n]:
# timestep_measure_event.append([timestep_to_datetime[clusters[e]["timestep"]], "{:.3f}".format(v),"{:.3f}".format(clusters[e]["chi"]),"{:.3f}".format(clusters[e]["alpha"]), e])
# latex_table.print_latex_table(data=timestep_measure_event, header=["timestep", "$\\beta$","$\chi$","$\\alpha$", "event"], alignment="llllp{12.0cm}", caption="$\\beta$, the std of peak heights", color="green")
chis = {event: clusters[event]["chi"] for event in clusters}
sorted_chis = sorted(chis.items(),
key=operator.itemgetter(1),
reverse=True)
timestep_measure_event = []
for e, v in sorted_chis[:n]:
timestep_measure_event.append([
timestep_to_datetime[clusters[e]["timestep"]], "{:.3f}".format(v),
"{:.3f}".format(clusters[e]["alpha"]),
"{:.3f}".format(clusters[e]["beta"]), e
])
latex_table.print_latex_table(
data=timestep_measure_event,
header=["timestep", "$\chi$", "$\\alpha$", "$\\beta$", "event"],
alignment="llllp{12.0cm}",
caption="$\chi$, highest peak intensity",
color="green")
return