def outlier_explanation_duration_by_dataset_chart(df,
time_unit,
microcluster_type='clu',
colors=('#047562',
'#D81B60'),
hatch=('\\', '-'),
figwidth=8,
figheight=6):
labels = df['stream name']
microcluster_col = f'{microcluster_type}-micoes execution time ({time_unit})'
microcluster_explainer_time = df[microcluster_col]
coin_col = f'coin execution time ({time_unit})'
coin_time = df[coin_col]
fig, ax = plt.subplots(figsize=(figwidth * 2, figheight))
# set width of bar
barWidth = 0.35
# Set position of bar on X axis
r1 = np.arange(len(labels))
r2 = [x + barWidth for x in r1]
# Make the plot
rects1 = ax.bar(r1,
microcluster_explainer_time,
color=colors[0],
width=barWidth,
edgecolor='white',
label=microcluster_col,
hatch=hatch[0])
rects2 = ax.bar(r2,
coin_time,
color=colors[1],
width=barWidth,
edgecolor='white',
label=coin_col,
hatch=hatch[1])
# Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_ylabel(f'duration ({time_unit})')
ax.set_title('stream name', y=-0.1)
ax.set_xticks(r1)
ax.set_xticklabels(labels)
ax.legend()
autolabel(rects1, ax)
autolabel(rects2, ax)
fig.tight_layout()
plt.show()
def microcluster_coin_percentage_chartv2(df,
colors=('#44AA99', '#188DC7'),
hatch=('\\', '/'),
figwidth=10,
figheight=6):
labels = df['stream name']
percentage_clu = df['clu-micoes over coin execution time (%)']
percentage_den = df['den-micoes over coin execution time (%)']
fig, ax = plt.subplots(figsize=(figwidth * 2, figheight))
# set width of bar
barWidth = 0.35
# Set position of bar on X axis
r1 = np.arange(len(labels))
r2 = [x + barWidth for x in r1]
mid = np.mean((r1, r2), axis=0)
# Make the plot
rects1 = ax.bar(r1,
percentage_clu,
color=colors[0],
width=barWidth,
edgecolor='white',
label='clu-micoes over COIN',
hatch=hatch[0])
rects2 = ax.bar(r2,
percentage_den,
color=colors[1],
width=barWidth,
edgecolor='white',
label='den-micoes over COIN',
hatch=hatch[1])
# Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_ylabel('LCOES over COIN execution time percentage (%)')
ax.set_title('Stream Names', y=-0.1)
ax.set_xticks(mid)
ax.set_xticklabels(labels)
ax.legend()
autolabel(rects1, ax)
autolabel(rects2, ax)
fig.tight_layout()
plt.show()
def microcluster_coin_percentage_by_dataset_chart(df,
microcluster_type='clu',
color='#CC6677',
figwidth=10,
figheight=6):
labels = df['stream name']
percentage_col = f'{microcluster_type}-micoes over coin execution time (%)'
percentage = df[percentage_col]
fig, ax = plt.subplots(figsize=(figwidth, figheight))
# set width of bar
barWidth = 0.5
# Set position of bar on X axis
r1 = np.arange(len(labels))
# Make the plot
rects1 = ax.bar(r1,
percentage,
color=color,
width=barWidth,
edgecolor='black')
# Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_ylabel(percentage_col)
ax.set_title('stream name', y=-0.1)
ax.set_xticks(r1)
ax.set_xticklabels(labels)
# ax.legend()
autolabel(rects1, ax)
fig.tight_layout()
plt.show()
def outlier_explanation_matched_by_dataset_chartv2(df,
time_unit,
colors=('#D81B60',
'#047562',
'#1E88E5'),
hatch=('-', '\\', '/'),
figwidth=8,
figheight=6):
labels = df['stream name']
df = df.round(decimals=2)
clu_matched = df[f'clu-micoes matched (%)']
den_matched = df[f'den-micoes matched (%)']
coin_matched = df[f'coin matched (%)']
fig, ax = plt.subplots(figsize=(figwidth * 2, figheight))
# set width of bar
barWidth = 0.3
# Set position of bar on X axis
r1 = np.arange(len(labels))
r2 = [x + barWidth for x in r1]
r3 = [x + barWidth for x in r2]
# Make the plot
rects1 = ax.bar(r1,
coin_matched,
color=colors[0],
width=barWidth,
edgecolor='white',
label='coin',
hatch=hatch[0])
rects2 = ax.bar(r2,
clu_matched,
color=colors[1],
width=barWidth,
edgecolor='white',
label='clu-micoes',
hatch=hatch[1])
rects3 = ax.bar(r3,
den_matched,
color=colors[2],
width=barWidth,
edgecolor='white',
label='den-micoes',
hatch=hatch[2])
# Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_ylabel(f'Accuracy (%)')
ax.set_title('Stream Names', y=-0.1)
ax.set_xticks(r2)
ax.set_xticklabels(labels)
ax.legend()
autolabel(rects1, ax)
autolabel(rects2, ax)
autolabel(rects3, ax)
fig.tight_layout()
plt.show()
def outlier_explanation_duration_by_window_chartv2(df,
stream_name,
time_unit,
colors=('#D81B60',
'#047562',
'#1E88E5'),
hatch=('-', '\\', '/'),
figwidth=8,
figheight=6):
labels = df['window size']
clu_time = np.ceil(df[f'clu-micoes execution time ({time_unit})'])
clu_time = clu_time.astype(int)
den_time = np.ceil(df[f'den-micoes execution time ({time_unit})'])
den_time = den_time.astype(int)
coin_time = np.ceil(df[f'coin execution time ({time_unit})'])
coin_time = coin_time.astype(int)
fig, ax = plt.subplots(figsize=(figwidth * 2, figheight))
# set width of bar
barWidth = 0.2
# Set position of bar on X axis
r1 = np.arange(len(labels))
r2 = [x + barWidth for x in r1]
r3 = [x + barWidth for x in r2]
# Make the plot
rects1 = ax.bar(r1,
coin_time,
color=colors[0],
width=barWidth,
edgecolor='white',
label='COIN',
hatch=hatch[0])
rects2 = ax.bar(r2,
clu_time,
color=colors[1],
width=barWidth,
edgecolor='white',
label='clu-micoes',
hatch=hatch[1])
rects3 = ax.bar(r3,
den_time,
color=colors[2],
width=barWidth,
edgecolor='white',
label='den-micoes',
hatch=hatch[2])
# Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_ylabel(f'Execution Time ({time_unit})')
ax.set_title(f'{stream_name} Window Size', y=-0.125)
ax.set_xticks(r2)
ax.set_xticklabels(labels)
ax.legend()
autolabel(rects1, ax)
autolabel(rects2, ax)
autolabel(rects3, ax)
fig.tight_layout()
plt.show()