def __init__(self, tlb_type, entry_num):
self.stats = stats.Statistics()
self.tlb_type = tlb_type
if tlb_type == Structure.FULLY_ASSOCIATIVE:
# a fully associative cache has multiple blocks in one set
self.set_num = entry_num
# We do not need an index for a fully-associative tlb/cache
self.set_index = 1
self.content = [tlb_entry.Entry()] * entry_num
elif tlb_type == Structure.DIRECT_MAP:
raise NotImplementedError(
"The direct map tlb is not implemented yet\n")
self.set_num = 1
# the index =
# the total number of entries
self.set_index = entry_num
self.init_content_set()
elif tlb_type == Structure.SET2_ASSOCIATIVE:
self.set_num = 2
# the index =
# the total number of entries / N (number of sets)
self.set_index = int(entry_num / 2)
self.init_content_set()
elif tlb_type == Structure.SET4_ASSOCIATIVE:
self.set_num = 4
self.set_index = int(entry_num / 4)
self.init_content_set()
else:
raise NotImplementedError("unimplemented tlb type")
def main():
print('-------------------------------------')
print('Machine Learning Foundation - Udacity')
print('Project #1: Explore US Bikeshare Data')
print('Student Name: Sugam Khetrapal')
print('-------------------------------------')
while True:
bkshare = BikeShare()
city_name, month_name, day = bkshare.user_input()
df = bkshare.load_city_file(city_name, month_name, day)
statistics = stats.Statistics(df)
statistics.time_stats()
statistics.station_stats()
statistics.trip_duration_stats()
statistics.user_stats()
statistics.table_stats(city_name)
start_again = input("Do you want to start again? (Y/N)").upper()
if start_again == 'N':
break
def setUp(self):
self.stats = stats.Statistics()
def tests(var,data,document,met,colors,run=''):
df=data.copy()
pk_reads=1
pk_assemblers=1
if var!="Ecoli strains":
if var!='Phred':
a = df.index.to_series().str.rsplit('_').str[-1].astype(float).sort_values()
df = df.reindex(index=a.index)
else:
d={"Genome":[],"Assembler":[],met:[]}
df_print=pd.DataFrame(data=d)
for col in df.T.columns:
gen=col.split('-')
gen=gen[0][4:]
for i,row in enumerate(df.T[col]):
df_print=df_print.append(pd.DataFrame([[gen,df.T.index[i],row]],columns=["Genome","Assembler",met]), ignore_index=True)
sns.swarmplot(x="Genome",y=met,hue="Assembler",data=df_print,palette="Set1")
plt.ylabel(met)
plt.xticks(rotation=90)
#5132068,4641652,5437407,4894879
#iai39,k12,o104,o83
fig = plt.gcf()
fig.set_size_inches(12, 8)
fig.savefig(out+met+'_swarm'+run+'.png', dpi=300)
fig.clear()
document.add_picture(out+met+'_swarm'+run+'.png', width=Inches(6.25))
table_docx(document,df)
ob=stats.Statistics(df,exp,var,met,"outputs_csv/",samples_int,run)
if var!="Ecoli strains" and run!='without_zscore':
if not os.path.isfile(out+met+run+".png"):
ob.scatter_plot(colors)
document.add_picture(out+met+run+'.png', width=Inches(6.25))
norms=ob.normality()
equal=[]
for item in df.values:
for each2 in item:
if each2 not in equal:
equal.append(each2)
#check if all the results are not the same. At least one need to be different
if len(equal)>1:
if norms[0] and norms[1]:
anova=ob.oneway()
'''anova_assemblers=anova[0].pvalue
anova_reads=anova[1].pvalue
if anova_assemblers <0.05 or anova_reads<0.05:
#ob.tukey_test()'''
k=ob.kruskal()
pk_assemblers=k[0].pvalue
pk_reads=k[1].pvalue
if pk_assemblers <0.05 or pk_reads<0.05:
nem=ob.nemenyi_test()
else:
k=ob.kruskal()
pk_assemblers=k[0].pvalue
pk_reads=k[1].pvalue
if pk_assemblers <0.05 or pk_reads<0.05:
nem=ob.nemenyi_test()
if run!='without_zscore' and var!="Ecoli strains":
ob.correlation_pearson()
document.add_picture(out+'corr_'+met+run+'.png', width=Inches(6.00))
try:
ob.linear_regression()
except:
pass
ob.results.close()
temp=open("outputs_csv/final_stats_"+met+""+run+".txt")
for line in temp:
document.add_paragraph(line.strip())
if pk_assemblers <0.05:
document.add_heading("Nemenyi Assemblers",level=5)
t = document.add_table(nem[0].shape[0]+1, nem[0].shape[1]+1,style='Light Grid Accent 1')
for j in range(nem[0].shape[-1]):
t.cell(0,j+1).text = nem[0].columns[j]
for i in range (len(nem[0].index)):
t.cell(i+1,0).text=nem[0].index[i]
# add the rest of the data frame
for i in range(nem[0].shape[0]):
for j in range(nem[0].shape[-1]):
t.cell(i+1,j+1).text = str(nem[0].values[i,j])
if pk_reads<0.05:
document.add_heading("Nemenyi samples",level=5)
t = document.add_table(nem[1].shape[0]+1, nem[1].shape[1]+1,style='Light Grid Accent 1')
for j in range(nem[1].shape[-1]):
t.cell(0,j+1).text = nem[1].columns[j]
for i in range (len(nem[1].index)):
t.cell(i+1,0).text=nem[1].index[i]
# add the rest of the data frame
for i in range(nem[1].shape[0]):
for j in range(nem[1].shape[-1]):
t.cell(i+1,j+1).text = str(nem[1].values[i,j])
if run!='without_zscore' and var!="Ecoli strains":
df.index=samples_int
if not os.path.isfile(out+met+'_plot'+run+'.png'):
df[var]=samples
df.plot(x=var, title=met)
plt.ylabel(met)
plt.xticks(rotation=90)
fig = plt.gcf()
fig.set_size_inches(12, 8)
fig.savefig(out+met+'_plot'+run+'.png', dpi=300)
fig.clear()
document.add_picture(out+met+'_plot'+run+'.png', width=Inches(6.25))
if not os.path.isfile(out+met+'_box'+run+'.png'):
df.plot.box(title=met,whis=3)
plt.ylabel(met)
plt.xticks(rotation=90)
fig = plt.gcf()
fig.set_size_inches(12, 8)
fig.savefig(out+met+'_box'+run+'.png', dpi=300)
fig.clear()
document.add_picture(out+met+'_box'+run+'.png', width=Inches(6.25))
return ob