def box_plot(y_positive, y_negative, y_neutral, name):
'''plot box plot takes in list of y variables that we want to plot'''
fig = go.Figure()
fig.add_trace(go.Box(y=y_positive,
name='positive'
))
fig.add_trace(go.Box(y=y_negative,
name='Negative'
))
fig.add_trace(go.Box(y=y_neutral,
name='Neutral'
))
fig.update_layout(
title={
'text': f"BoxPlot of sentiments of {name}",
'y': 0.99,
'x': 0.5,
'xanchor': 'center',
'yanchor': 'top',
'font': dict(
family='Rockwell',
color='dim gray',
size=30),
},
#template='plotly_dark',
xaxis=dict(
showgrid=False
),
yaxis=dict(
showgrid=False
))
return(offl.plot(fig, show_link=False, output_type="div", include_plotlyjs=False))
def get_figure1(df):
fig = go.Figure()
fig.add_trace(
go.Box(x=df['sodium_dv'].dropna() * 100,
name='Sodium',
boxmean=True))
fig.add_trace(
go.Box(x=df['totalfat_dv'].dropna() * 100,
name='Total Fat',
boxmean=True))
fig.add_trace(
go.Box(x=df['sugar'].dropna(), name='Sugar', boxmean=True))
fig.update_layout(barmode='stack')
fig.update_traces(opacity=0.75)
fig['layout']['xaxis'].update(range=[0, 100])
fig.update_layout(
xaxis_title="Daily Value (%)",
margin=dict(
l=0,
r=0,
b=50,
t=50,
),
title={
'text': 'Distribution of Select Nutrients (Stores Combined)',
'x': 0.5,
'xanchor': 'center',
'yanchor': 'top'
},
height=400)
return to_html(fig, include_plotlyjs=False, full_html=False)
def addBox(self, column):
if self.orientation == 'V':
self.fig.add_trace(go.Box(y=column))
elif self.orientation == 'H':
self.fig.add_trace(go.Box(x=column))
else:
self.fig.add_trace(go.Box(y=column))
def figure_box_hist(df, group, linear, bar):
""" Update box plot and histogram in comparison tab depending on drop downs """
df_linear = df[df['type'] == linear]
df_bar = df[df['type'] == bar]
name = df[group].unique()
colors = px.colors.qualitative.Plotly
fig_histogram = make_subplots(rows=1, cols=2, subplot_titles=(linear, bar))
fig_box_plot = make_subplots(rows=1, cols=2, subplot_titles=(linear, bar))
for name, i in enumerate(name):
df_l = df_linear[df_linear[group] == i]
df_b = df_bar[df_bar[group] == i]
i = str(i)
fig_histogram.add_trace(go.Histogram(x=df_l['Value'], name=i, showlegend=False, legendgroup=i,
marker_color=colors[name]), row=1, col=1)
fig_histogram.add_trace(go.Histogram(x=df_b['Value'], name=i, legendgroup=i, marker_color=colors[name]),
row=1, col=2)
fig_box_plot.add_trace(go.Box(y=df_l['Value'],name=i, showlegend=False, legendgroup=i,
marker_color=colors[name]), row=1, col=1)
fig_box_plot.add_trace(go.Box(y=df_b['Value'], name=i, legendgroup=i, marker_color=colors[name]), row=1, col=2)
# Overlay both histograms
fig_histogram.update_layout(barmode='stack')
fig_histogram.update_layout(template='plotly_white')
fig_box_plot.update_layout(template='plotly_white')
return fig_box_plot, fig_histogram
def plot_weather(process_dict):
#plotting
#boxplot
fig1a = go.Figure()
fig1a.add_trace(go.Box(y=process_dict['overallTs'], name="Temperature"))
fig1a.add_trace(
go.Box(y=process_dict['overallRFTs'], name="Real Feel Temperature"))
fig1a.update_layout(
title=
f"Boxplot comparison of Temperature and Real Feel Temperature on {process_dict['Date']} for the past {process_dict['File']}",
xaxis_title='Variable',
yaxis_title='Temperature (°C)')
fig1a.show()
#bar graph
fig1b = go.Figure()
fig1b.add_trace(
go.Bar(x=process_dict['WeatherText'],
y=process_dict['WeatherFreq'],
name="Weather"))
fig1b.update_layout(
title=
f"Frequency comparison of WeatherText on {process_dict['Date']} for the past {process_dict['File']}",
xaxis_title='WeatherText Category',
yaxis_title='Count')
fig1b.show()
def boxplot_through_the_day(plot_df):
"""Returns a plot with the boxplot of the estimations through the day"""
fig = go.Figure()
fig.add_trace(
go.Box(
x=plot_df['departure_time'].dt.hour,
y=plot_df['trip_time_estimate'].dt.total_seconds() / 60,
name='running time estimate',
marker=dict(color='red', ),
))
fig.add_trace(
go.Box(
x=plot_df['departure_time'].dt.hour,
y=plot_df['trip_time'].dt.total_seconds() / 60,
name='running time',
marker=dict(color='green', ),
))
#Edit the layout
fig.update_layout(
title='Estimated running times and running times through the day',
xaxis_title='Hour of the day',
yaxis_title='Running time (minutes)')
return fig
def get_boxplots(df, ts_df):
df_copy = df.copy()
ts_df_copy = ts_df.copy()
df_copy["text"] = df["id"].astype(str) + " si(" + df["si"].astype(str) + "), length(" + df["length"].astype(str) \
+ "), area(" + df["area"].astype(str) + ")"
ts_df_copy["text"] = ts_df["id"].astype(str) + " si(" + ts_df["si"].astype(str) \
+ "), length(" + ts_df["length"].astype(str) + "), area(" + ts_df["area"].astype(str) + ")"
fig = make_subplots(rows=1, cols=5)
fig.add_trace(go.Box(y=list(df_copy["si"]),
name="si (event)",
text=df_copy["text"]),
row=1,
col=1)
fig.add_trace(go.Box(y=list(ts_df_copy["si"]),
name="si (timeseries)",
text=ts_df_copy["text"]),
row=1,
col=2)
fig.add_trace(go.Box(y=list(df_copy["length"]),
name="length",
text=df_copy["text"]),
row=1,
col=3)
fig.add_trace(go.Box(y=list(df_copy["area"]),
name="area",
text=df_copy["text"]),
row=1,
col=4)
fig.add_trace(go.Box(y=list(ts_df_copy["meanPre"]),
name="meanPre",
text=ts_df_copy["text"]),
row=1,
col=5)
return fig
def visualize_age_relations(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "age_relations", dbutils)
display_positive = dataframe_pd["positive"]
display_negative = dataframe_pd["negative"]
display_age = dataframe_pd["age"]
rec_age_fig = make_subplots(
rows=1,
cols=2,
subplot_titles=("Positive test/age coefficient correlation",
"Negative test/age coefficient correlation"))
rec_age_fig.add_trace(go.Box(x=display_positive,
y=display_age,
name="Positive",
marker_color=color_scheme.color_300),
row=1,
col=1)
rec_age_fig.add_trace(go.Box(x=display_negative,
y=display_age,
name="Negative",
marker_color=color_scheme.color_700),
row=1,
col=2)
rec_age_fig.update_traces(boxpoints='all')
rec_age_fig.update_layout(
title_text="Subplots of age in relation a positive/negative test result"
)
rec_age_fig.show()
return None
def box_wp():
layout = dict(title="", template="plotly_white", showlegend=False)
fig = go.Figure(layout=layout)
# whole pancreas protein
try:
wpp_wt = [
whole_pancreas_pro_wt_1, whole_pancreas_pro_wt_2,
whole_pancreas_pro_wt_3, whole_pancreas_pro_wt_4
]
wpp_g12d = [
whole_pancreas_pro_g12d_1, whole_pancreas_pro_g12d_2,
whole_pancreas_pro_g12d_3
]
fig.add_trace(
go.Box(x=['WT', 'WT', 'WT', 'WT'],
y=wpp_wt,
name='WT',
marker_color='lightseagreen',
boxpoints='all',
pointpos=0))
fig.add_trace(
go.Box(x=['G12D', 'G12D', 'G12D'],
y=wpp_g12d,
name='G12D',
marker_color='lightseagreen',
boxpoints='all',
pointpos=0))
except:
fig.add_trace(go.Box(y=[], name='Whole pancreas WT'))
fig.add_trace(go.Box(y=[], name='Whole pancreas G12D'))
fig.update_layout(xaxis={'title': 'Whole pancreas'},
yaxis={'title': None})
plot_div = plot(fig, output_type='div', include_plotlyjs=False)
return plot_div
def box_ce():
layout = dict(title="", template="plotly_white", showlegend=False)
fig = go.Figure(layout=layout)
# scraped colon protein
try:
scp_wt = [
scraped_colon_pro_fabp_1, scraped_colon_pro_fabp_2,
scraped_colon_pro_fabp_4, scraped_colon_pro_fabp_5
]
scp_g12d = [
scraped_colon_pro_g12d_1, scraped_colon_pro_g12d_2,
scraped_colon_pro_g12d_3, scraped_colon_pro_g12d_4
]
fig.add_trace(
go.Box(x=['WT', 'WT', 'WT', 'WT'],
y=scp_wt,
name='WT',
marker_color='#636EFA',
boxpoints='all',
pointpos=0))
fig.add_trace(
go.Box(x=['G12D', 'G12D', 'G12D', 'G12D'],
y=scp_g12d,
name='G12D',
marker_color='#636EFA',
boxpoints='all',
pointpos=0))
except:
fig.add_trace(go.Box(y=[], name='Colonic epithelium WT'))
fig.add_trace(go.Box(y=[], name='Colonic epithelium G12D'))
fig.update_layout(xaxis={'title': 'Colonic epithelium'},
yaxis={'title': None})
plot_div = plot(fig, output_type='div', include_plotlyjs=False)
return plot_div
def plotBoxPDF(listOfDictonaries, opFile):
pathToDir, opFileName = os.path.split(opFile)
# First check if the path entered by the user is correct or not.
if os.path.exists(os.path.abspath(pathToDir)):
pathToOutputDir = os.path.join(os.path.abspath(pathToDir), opFileName)
# Store the output file name and use it in the with open loop
fileName = str(pathToOutputDir) + ".pdf"
fig = go.Figure()
for i in listOfDictonaries:
y_axis = i['tmpMed']
hopName = "Hop " + str(i['hop'])
check_flag = i['avg']
if check_flag != 0:
fig.add_trace(go.Box(y=y_axis, name=hopName))
fig.write_image(fileName)
print("Program excuted successfully and file is stored at:", fileName)
else:
print("The path entered is not a valid path:", pathToDir)
print("Creating the output file at:", os.getcwd())
fileName = opFileName + ".pdf"
fig = go.Figure()
for i in listOfDictonaries:
y_axis = i['tmpMed']
hopName = "Hop " + str(i['hop'])
check_flag = i['avg']
if check_flag != 0:
fig.add_trace(go.Box(y=y_axis, name=hopName))
fig.write_image(fileName)
def BoxPlotCoEvolvedFiles(co_evolved, not_co_evolved):
fig = go.Figure()
fig.add_trace(
go.Box(y=co_evolved,
name="Co-Evolved Files",
marker_color="cornflowerblue"))
fig.add_trace(
go.Box(y=not_co_evolved,
name="Not Co-Evolved Files",
marker_color="crimson"))
fig.update_layout(
width=900,
height=550,
yaxis_title='Distribution of Co-Evolved Files in Commits',
# legend=dict(
# yanchor="top",
# y=0.99,
# xanchor="right",
# x=0.99
# )
showlegend=False)
fig.show()
def grouped_box_plot(data,
cat_col,
num_cols,
title=None,
y_title=None,
x_title=None,
points=True):
"""
Create a box plot for a continous variable grouped by a categorical variable
:param data:
:type data:
:param cat_col:
:type cat_col:
:param num_col:
:type num_col:
:return:
:rtype:
"""
groups = data.groupby(cat_col)
fig = go.Figure()
for variable in num_cols:
trace_x = []
trace_y = []
for idx, group in groups:
y = list(group[variable][group[variable].notnull()])
x = [idx for i in range(len(y))]
trace_y += y
trace_x += x
# Wrap the plot titles
if len(variable) > 30:
display_var = "<br>".join(textwrap.wrap(variable, width=30))
else:
display_var = variable
if points:
fig.add_trace(
go.Box(y=trace_y,
x=trace_x,
name=display_var,
boxpoints='all',
jitter=0.2))
else:
fig.add_trace(go.Box(y=trace_y, x=trace_x, name=display_var))
fig.update_layout(title=title,
title_x=0.5,
yaxis_title=y_title,
xaxis_title=cat_col,
boxmode='group',
width=1000)
if x_title:
fig.update_layout(xaxis_title=x_title, )
legend = dict(yanchor="top", y=0.99, x=1.02)
fig.update_layout(legend=legend)
fig.write_image("grouped_box_plot.png")
fig.show()
def main(dataframe_files):
"""
:param dataframe_files:
:return:
"""
df = pd.concat((pd.read_csv(f) for f in dataframe_files))
for sample_id in sorted(df['Sample'].unique()):
fig = go.Figure()
for assembler in sorted(df['Assembler'].unique()):
contigs = df['Contig Len'][df['Assembler'] == assembler]
mapped_contigs = df['Contig Len'][(df['Mapped'] != 'Unmapped')
& (df['Assembler'] == assembler)]
print(','.join([
assembler,
f'{len(mapped_contigs)} ({(len(mapped_contigs) / len(contigs)) * 100:.2f}%)',
f'{sum(mapped_contigs)} ({(sum(mapped_contigs) / sum(contigs)) * 100:.2f}%)'
]))
# mapped contigs as boxplots
fig.add_trace(
go.Box(x=df['Contig Len'][(df['Mapped'] != 'Unmapped')
& (df['Assembler'] == assembler)],
name=assembler,
boxpoints='outliers',
boxmean=False,
fillcolor='#D3D3D3',
line=dict(color='#000000')))
# unmapped contigs as scatter-like plot (boxplot showing only the underlying data)
fig.add_trace(
go.Box(x=df['Contig Len'][(df['Mapped'] == 'Unmapped')
& (df['Assembler'] == assembler)],
name=assembler,
boxpoints='all',
pointpos=0,
marker=dict(color='rgba(178,37,34,0.7)'),
line=dict(color='rgba(0,0,0,0)'),
fillcolor='rgba(0,0,0,0)'))
fig.update_layout(
showlegend=False,
xaxis_type="log",
xaxis_title="Contig size (Log bp)",
title=
"Contig size distribution per assembler (contigs over 1000 bp)",
plot_bgcolor='rgb(255,255,255)',
xaxis=dict(zeroline=False, gridcolor='#DCDCDC'))
plot(fig,
filename='{}_contig_size_distribution.html'.format(sample_id),
auto_open=False)
fig.write_json(
file='{}_contig_size_distribution.json'.format(sample_id))
def update_tumor_volume(input_):
traces = []
print(input_)
if input_ is None:
drug_name = 'Placebo'
traces.append(
go.Box(x=merged_df[merged_df['Drug Regimen'] ==
drug_name]['Timepoint'],
y=merged_df[merged_df['Drug Regimen'] == drug_name]
['Tumor Volume (mm3)'],
marker_color=drug_colors[drug_name],
name=drug_name))
else:
# drug_list = input_['points'][1]['Drug Regimen']
drug_list = []
for drug_name in drug_list:
traces.append(
go.Box(x=merged_df[merged_df['Drug Regimen'] ==
drug_name]['Timepoint'],
y=merged_df[merged_df['Drug Regimen'] == drug_name]
['Tumor Volume (mm3)'],
marker_color=drug_colors[drug_name],
name=drug_name,
customdata=[drug_name]))
return {
'data':
traces,
'layout':
dict(xaxis={
'title': 'time point',
},
yaxis={
'title':
'tumor volume mm3',
'range': [
merged_df['Tumor Volume (mm3)'].min(),
merged_df['Tumor Volume (mm3)'].max()
],
'showgrid':
False
},
boxmode='group',
autosize=False,
paper_bgcolor=colors['chart-background'],
plot_bgcolor=colors['chart-background'],
margin={
'l': 40,
'b': 40,
't': 10,
'r': 10
},
legend={
'x': 0.99,
'y': 1
},
clickmode='event+select')
}
def makestatsGraph(date=None):
fig = go.Figure()
tmp = df2[df2.Date == df2.Date.max()].copy()
fig.add_trace(go.Box(y=np.log(tmp.Mort), name="Mortality"))
fig.add_trace(go.Box(y=np.log(tmp.DP100K + 1), name="Deaths per 100K"))
fig.add_trace(go.Box(y=np.log(tmp.CP100K + 1), name="Cases per 100K"))
fig.update_layout(title="Box Plots",
title_x=0.5,
yaxis=dict(title="Logarithmic scale"))
return fig
def C(selectinfo, checkinfo):
traces = []
drug_list = []
if selectinfo:
for drug in selectinfo['points']:
if drug['customdata'][0] not in drug_list:
drug_list.append(drug['customdata'][0])
if drug_list:
for drug_name in drug_list:
df_box = mdf[mdf['Drug Regimen'] == drug_name]
traces.append(
go.Box(
x=df_box['Timepoint'],
y=df_box['Tumor Volume (mm3)'],
opacity=0.6,
name=drug_name,
marker=dict(color=drug_colors[drug_name]),
))
else:
for drug_name in checkinfo:
df_box = mdf[mdf['Drug Regimen'] == drug_name]
traces.append(
go.Box(
x=df_box['Timepoint'],
y=df_box['Tumor Volume (mm3)'],
opacity=0.6,
name=drug_name,
marker=dict(color=drug_colors[drug_name]),
))
return {
'data':
traces,
'layout':
dict(boxmode="group",
xaxis={'title': 'timepoint'},
yaxis={
'title': 'tumor Volume (mm3)',
'showgrid': False
},
autosize=False,
paper_bgcolor=colors['chart-background'],
plot_bgcolor=colors['chart-background'],
margin={
'l': 40,
'b': 40,
't': 10,
'r': 10
},
showlegend=True,
legend_traceorder="reversed",
legend={
'x': 1,
'y': 1
})
}
def box_plot():
np.random.seed(1)
y0 = np.random.randn(50) - 1
y1 = np.random.randn(50) + 1
fig = go.Figure()
fig.add_trace(go.Box(y=y0))
fig.add_trace(go.Box(y=y1))
fig.update_layout(title='Box Plot')
plot_div = plot(fig, output_type='div', include_plotlyjs=False)
return plot_div
def buildgraph_ex3(data1, data2):
fig1 = go.Figure()
fig1.add_trace(
go.Scatter(mode='markers', x=data1, marker=dict(color='Blue')))
fig1.add_trace(
go.Scatter(mode='markers', x=data2, marker=dict(color='Red')))
fig1.add_shape(type="line",
x0=np.mean(data1),
y0=1,
x1=np.mean(data1),
y1=100,
line=dict(color="Blue", width=4, dash="dashdot"))
fig1.add_shape(type="line",
x0=np.mean(data2),
y0=1,
x1=np.mean(data2),
y1=100,
line=dict(color="Red", width=4, dash="dashdot"))
fig1.update_layout(barmode='overlay')
fig1.update_layout(xaxis_range=[-10, 10])
fig1.update_traces(opacity=0.7)
fig1.layout.update(showlegend=False)
fig1.update_layout(title_text='Scatter Plot')
graph1 = dcc.Graph(figure=fig1)
fig2 = go.Figure()
fig2.add_trace(go.Histogram(x=data1, name='Group 1'))
fig2.add_trace(go.Histogram(x=data2, name='Group 2'))
fig2.update_layout(barmode='overlay')
fig2.update_traces(opacity=0.7)
fig2.update_layout(title_text='Histogram')
graph2 = dcc.Graph(figure=fig2)
fig3 = go.Figure()
fig3.add_trace(go.Box(y=data1, name='Group 1'))
fig3.add_trace(go.Box(y=data2, name='Group 2'))
fig3.update_layout(title_text='Boxplot')
graph3 = dcc.Graph(figure=fig3)
mult = sc.t.ppf((1 + 0.95) / 2, len(data1) - 1)
fig4 = go.Figure()
fig4.add_trace(
go.Bar(name='Control',
x=['Group 1', 'Group 2'],
y=[np.mean(data1), np.mean(data2)],
error_y=dict(type='data',
array=[sc.sem(data1) * mult,
sc.sem(data2) * mult])))
fig4.update_layout(title_text='Bargraph with Errorbars')
graph4 = dcc.Graph(figure=fig4)
return graph1, graph2, graph3, graph4
def get_log_residual_plot():
#meta_pre = ['meta_pre_g1', 'meta_pre_g2', 'meta_pre_g3', 'meta_pre_g4', 'meta_pre_g5', 'meta_pre_g6', 'meta_pre_g7', 'meta_pre_g8']
#video_pre = ['video_pre_g1', 'video_pre_g2', 'video_pre_g3', 'video_pre_g4', 'video_pre_g5', 'video_pre_g6', 'video_pre_g7', 'video_pre_g8']
current_dir = os.path.dirname(os.path.realpath(__file__))
data_file = os.path.join(current_dir, 'data', 'summary_final_result.csv')
#csv = np.genfromtxt(data_file, delimiter=None, comments='#')
g8_result_final = pd.read_csv(data_file)
meta_res_log = [
'meta_res_log_g1', 'meta_res_log_g2', 'meta_res_log_g3',
'meta_res_log_g4', 'meta_res_log_g5', 'meta_res_log_g6',
'meta_res_log_g7', 'meta_res_log_g8'
]
video_res_log = [
'video_res_log_g1', 'video_res_log_g2', 'video_log_res_g3',
'video_res_log_g4', 'video_res_log_g5', 'video_res_log_g6',
'video_res_log_g7', 'video_res_log_g8'
]
#g8_result_final = pd.read_csv('summary_final_result.csv')
N = 8
c = ['hsl(' + str(h) + ',50%' + ',50%)' for h in np.linspace(0, 360, N)]
d = ['hsl(' + str(h) + ',30%' + ',30%)' for h in np.linspace(0, 360, N)]
group = [
'Group1', 'Group2', 'Group3', 'Group4', 'Group5', 'Group6', 'Group7',
'Group8'
]
fig = go.Figure()
for i in range(int(N)):
fig.add_trace(
go.Box(x=g8_result_final[meta_res_log[i]],
y=[group[i]] * len(g8_result_final[meta_res_log[i]]),
name='Logarithm of meta residuals of Group '
'{}'.format(i + 1),
marker_color=c[i]))
fig.add_trace(
go.Box(x=g8_result_final[video_res_log[i]],
y=[group[i]] * len(g8_result_final[video_res_log[i]]),
name='Logarithm of video residuals of Group '
'{}'.format(i + 1),
marker_color=d[i]))
fig.update_layout(xaxis=dict(title='The absolute logarithm of residuals',
zeroline=False),
boxmode='group')
fig.update_traces(orientation='h') # horizontal box plots
#fig.show()
plot_div = plot(fig,
output_type='div',
filename='Logarithm of residuals for 8 groups')
return plot_div
def box_plot(data,Analysis_column):
st.header('**Data Analysis**')
box_train = data
box_train['ds'] = pd.to_datetime(box_train['ds'])
box_train['Hourly'] = box_train['ds'].dt.hour
box_train['Daily'] = box_train['ds'].dt.day
box_train['Monthly'] = box_train['ds'].dt.month
box_train['Yearly'] = box_train['ds'].dt.year
daily = box_train.groupby(['Daily'],as_index=False)['y'].mean()
Monthly = box_train.groupby(['Monthly'],as_index=False)['y'].mean()
Yearly = box_train.groupby(['Yearly'],as_index=False)['y'].mean()
fig = make_subplots(rows=3, cols=2)
fig.add_trace(
go.Scatter(x=daily.index,y=daily['y'],
name='Daily Trend'),
row = 1, col=1
)
fig.add_trace(
go.Box(x=box_train['Daily'],y=box_train['y'],
name='Daily Trend BoxPlot'),
row = 1, col=2
)
fig.add_trace(
go.Scatter(x=Monthly.index,y=Monthly['y'],
name='Monthly Trend'),
row = 2, col=1
)
fig.add_trace(
go.Box(x=box_train['Monthly'],y=box_train['y'],
name='Monthly Trend BoxPlot'),
row = 2, col=2
)
fig.add_trace(
go.Scatter(x=Yearly['Yearly'],y=Yearly['y'],
name='Yearly Trend'),
row = 3, col=1
)
fig.add_trace(
go.Box(x=box_train['Yearly'],y=box_train['y'],
name='Yearly Trend BoxPlot'),
row = 3, col=2
)
fig.update_xaxes(title_text = 'Daily Time Period', row=1, col =1)
fig.update_xaxes(title_text = 'Monthly Time Period', row=2, col =1)
fig.update_xaxes(title_text = 'Yearly Time Period', row=3, col =1)
fig.update_yaxes(title_text = Analysis_column, row=1, col =1)
fig.update_yaxes(title_text = Analysis_column, row=2, col =1)
fig.update_yaxes(title_text = Analysis_column, row=3, col =1)
fig.update_layout(
plot_bgcolor = '#FFF', autosize=False, width=1100, height=1000,
)
st.plotly_chart(fig)
def update_statistics_charts(project, subject, gender, race, age):
if project == 'all':
project = data.project.unique()
else:
project = [project]
if not subject:
return px.box(), px.box()
elif 'all' in subject:
subject = data.person.unique()
else:
subject = subject
if gender == 'all':
gender = data.gender.unique()
else:
gender = [gender]
if not race:
return px.box(), px.box()
elif 'all' in race:
race = data.race.unique()
else:
race = race
mask = ((data.project.isin(project)) & (data.person.isin(subject)) &
(data.gender.isin(gender)) & (data.race.isin(race)) &
((data.age <= age[1]) & (data.age >= age[0])))
filtered_data = data.loc[mask, :]
mape_chart_figure = {
"data": [
go.Box(x=filtered_data['mape'],
y=filtered_data['activity'],
orientation='h')
],
"layout":
go.Layout(title='MAPE vs Activity'),
}
da_chart_figure = {
"data": [
go.Box(x=filtered_data['da'],
y=filtered_data['activity'],
orientation='h')
],
"layout":
go.Layout(title='Data Availability vs Activity'),
}
return mape_chart_figure, da_chart_figure
def plot_cps():
Y_entrees_cps = []
Y_breakfast_entrees_cps = []
Y_ventrees_cps = []
Y_desserts_cps = []
get_calories_per_serving(12, Y_entrees_cps)
get_calories_per_serving(2, Y_breakfast_entrees_cps)
get_calories_per_serving(13, Y_ventrees_cps)
get_calories_per_serving(17, Y_desserts_cps)
fig = go.Figure()
fig.add_trace(
go.Box(x=Y_entrees_cps,
marker_color=primary_colors[0],
name='Entrees',
boxpoints='outliers'))
fig.add_trace(
go.Box(x=Y_breakfast_entrees_cps,
marker_color=primary_colors[1],
name='Breakfast Entrees',
boxpoints='outliers'))
fig.add_trace(
go.Box(x=Y_ventrees_cps,
marker_color=primary_colors[2],
name='Vegetarian Entrees',
boxpoints='outliers'))
fig.add_trace(
go.Box(x=Y_desserts_cps,
marker_color=primary_colors[3],
name='Desserts',
boxpoints='outliers'))
fig.update_layout(
title='Calories per Serving of HUDS Entrees and Desserts',
xaxis=dict(tickmode='linear',
tick0=0,
dtick=5,
title={'text': 'Calories per Ounce or Fluid Ounce'}),
xaxis_range=[0, 130],
yaxis={'title': {
'text': 'Food Categories'
}},
legend={'title': {
'text': 'Groups'
}},
template=theme_hodp)
fig.show()
def update_box(hoverData1,hoverData2, var1,var2, type):
"""Updates box plot in response to change in granularity, vars,
or which precinct user is hovering over"""
if type == 'Precincts':
try:
precinctkey1 = hoverData1['points'][0]['location']
precinctkey2 = hoverData2['points'][0]['location']
except TypeError:
precinctkey1 = 270402
precinctkey2 = 270402
merged1 = aggregate('Precincts', var1,resultscatalog)
merged2 = aggregate('Precincts',var2,resultscatalog)
df1 = merged1[['PrecinctKey',var1]]
df2 = merged2[['PrecinctKey',var2]]
index1 = df1.index[df1['PrecinctKey'] == precinctkey1].tolist()
index2 = df2.index[df2['PrecinctKey'] == precinctkey2].tolist()
box =make_subplots(rows=1,cols=2)
box.add_trace(go.Box(y=df1[var1], boxpoints='all',
name=var1,text=df1['PrecinctKey'],hoverinfo='y+text',selectedpoints=index1,
hovertemplate="PrecinctKey: %{text} | Value: %{y}<extra></extra>"),
row=1,col=1)
box.add_trace(go.Box(y=df2[var2], boxpoints='all',
name=var2,text=df2['PrecinctKey'],hoverinfo='y+text',selectedpoints=index2,
hovertemplate="PrecinctKey: %{text} | Value: %{y}<extra></extra>"),
row=1,col=2)
box.update_layout(showlegend=False)
else: #for zipcode mappings
try:
zip1 = hoverData1['points'][0]['location']
zip2 = hoverData2['points'][0]['location']
except TypeError:
zip1 = 76543
zip2 = 76543
merged1 = aggregate('Zipcodes',var1,resultscatalog)
merged2 = aggregate('Zipcodes',var2,resultscatalog)
df1 = merged1[['zipcode',var1]]
df2 = merged2[['zipcode',var2]]
index1 = df1.index[df1['zipcode'] == zip1].tolist()
index2 = df2.index[df2['zipcode'] == zip2].tolist()
box =make_subplots(rows=1,cols=2)
box.add_trace(go.Box(y=df1[var1], boxpoints='all',
name=var1,text=df1['PrecinctKey'],hoverinfo='y+text',selectedpoints=index1,
hovertemplate="PrecinctKey: %{text} | Value: %{y}<extra></extra>"),
row=1,col=1)
box.add_trace(go.Box(y=df2[var2], boxpoints='all',
name=var2,text=df2['PrecinctKey'],hoverinfo='y+text',selectedpoints=index2,
hovertemplate="PrecinctKey: %{text} | Value: %{y}<extra></extra>"),
row=1,col=2)
box.update_layout(showlegend=False)
return box
def box_plot(df, xaxis, yaxis, boxpt, boxaxis):
fig = go.Figure()
if boxpt=='False': boxpt=False
if boxaxis == 'xaxis':
fig.add_trace(go.Box(x=df[xaxis], boxpoints=boxpt, name=f'{xaxis}', marker_color='#636EFA'))
elif boxaxis == 'yaxis':
fig.add_trace(go.Box(x=df[yaxis], boxpoints=boxpt, name=f'{yaxis}', marker_color='#19D3F3'))
else:
fig.add_trace(go.Box(x=df[xaxis], boxpoints=boxpt, name=f'{xaxis}', marker_color='#636EFA'))
fig.add_trace(go.Box(x=df[yaxis], boxpoints=boxpt, name=f'{yaxis}', marker_color='#19D3F3'))
fig.update_xaxes(showgrid=False, zeroline=False)
fig.update_yaxes(showgrid=False, zeroline=False)
fig.update_layout(plot_bgcolor='rgb(255,255,255)', height=450, width=1200)
return fig
def plot_boxplots(year_range, month_range, si_range, area_range, map_size_radio_items, hours_range, country_list):
tmp = filter_events(year_range, month_range, si_range, area_range, map_size_radio_items, hours_range, country_list)
# TODO CODE
# Histogram with all outliers combined as e.g. >30 hours
# return figure
fig = make_subplots(rows=1,cols=4)
fig.append_trace(go.Box(y=tmp.event_si, name="Severity",marker=dict(color=color_set[0])),row=1,col=1)
fig.append_trace(go.Box(y=tmp.event_area, name="Area",marker=dict(color=color_set[1])),row=1,col=2)
fig.append_trace(go.Box(y=tmp.event_pre, name="Precipitation",marker=dict(color=color_set[2])),row=1,col=3)
fig.append_trace(go.Box(y=tmp.event_length, name="Duration",marker=dict(color=color_set[3])),row=1,col=4)
fig.update_layout(title="Dispertion of Attributes",legend=dict(orientation="h"))
return fig
def boxplot(df,
title='Box-Plot',
out_path=None,
layout_width=None,
layout_height=None,
numeric_only=True):
data = []
columns = [
k for k, v in df.dtypes.items() if 'float' in str(v) or 'int' in str(v)
] if numeric_only else df.columns
for column in columns:
data.append(
go.Box(
y=df[column],
boxpoints='outliers', # all, outliers, suspectedoutliers
boxmean=True))
max_col = 2
subplot_titles = [f'{x.lower()}' for x in columns]
plot_subplots(data,
max_col,
title,
subplot_titles=subplot_titles,
out_path=out_path,
layout_width=layout_width,
layout_height=layout_height)
def plot_corr_box_plot(self, x_axis_column_name, y_axis_column_name,
scale):
fig = make_subplots(rows=1, cols=1)
shared_bins_cond = None
for plot_df, method_name, j in zip(self.plot_dfs, self.method_names,
range(len(self.plot_dfs))):
plot_df = filter_by_scale(scale, plot_df)
# plot_df = plot_df[(np.log2(plot_df[x_axis_column_name]+1) >=1) & (np.log2(plot_df[y_axis_column_name]+1) >= 1)]
df, shared_bins_cond = prepare_corr_box_plot_data(
np.log2(plot_df[y_axis_column_name] + 1),
np.log2(plot_df[x_axis_column_name] + 1), shared_bins_cond)
fig.add_trace(go.Box(x=df['true_abund'],
y=df['estimated_abund'],
name=method_name),
col=1,
row=1)
fig.update_xaxes(title_text='Log2(True TPM+1)')
fig.update_yaxes(title_text='Log2(Estimated TPM+1)')
fig.update_layout(boxmode="group",
autosize=False,
showlegend=True,
width=fig_size['square']['width'],
height=fig_size['square']['height'],
template=themes['large_single'])
return fig
def plot_normalized_demand(data):
"""
Plot Tempearture Normalized Demand based on Heating and Cooling degree days
Input: Data from 2010 - 2020
Returns: Figure
"""
temp_df = data[["AIL_DEMAND", "Degree_days"]].copy()
temp_df = temp_df.groupby(by=temp_df.index.date).sum()
temp_df.index = pd.to_datetime(temp_df.index)
temp_df['Normalized_AIL_Demand'] = temp_df.AIL_DEMAND / temp_df.Degree_days
fig = go.Figure()
for year in range(2010, 2021):
fig.add_trace(
go.Box(
x=temp_df[temp_df.index.year == year].index.year,
y=temp_df['Normalized_AIL_Demand'][temp_df.index.year == year],
name=str(year)))
fig.update_layout(
title=
"Distribition of Daily Demand per Heating/Cooling Degree Days (Base = 18 Celsius) "
)
return fig
def box(df, title='Box', color=None,
out_path=None, max_col=2, layout_kwargs={}, to_image=False,
box_kwargs={}):
columns = df.select_dtypes(include='number')
columns = [x for x in columns if x != color]
data_groups = []
groups = [1] if color is None else df[color].unique()
for column in columns:
data = []
colors = DEFAULT_PLOTLY_COLORS
for index, group in enumerate(groups):
data.append(go.Box(
y=df[column] if color is None else df[df[color] == group][column],
name=column if color is None else group,
marker={'color': colors[index % len(colors)]},
**box_kwargs
))
data_groups.append(data)
if color is None:
layout_kwargs['showlegend'] = False
datagroups_subplots(data_groups, max_col=max_col, title=title, out_path=out_path,
yaxis_titles=[] if color is None else columns,
layout_kwargs=layout_kwargs, to_image=to_image)
