def plot_gps_clusters(ds, user_id:str, zoom=5):
"""
Plots GPS coordinates
Args:
ds (DataStream): datastream object
user_id (str): uuid of a user
zoom: min 0 and max 100, zoom map
"""
pdf = ds_to_pdf(ds, user_id)
marker_list = []
center = None
for index, row in pdf.iterrows():
if center is None:
center = [row["latitude"], row["longitude"]]
marker_list.append(Marker(location=(row["latitude"], row["longitude"])))
m = Map(center=(center), zoom=zoom)
marker_cluster = MarkerCluster(
markers=(marker_list)
)
m.add_layer(marker_cluster)
return m
def plot_gantt(ds, user_id):
"""
Args:
ds:
user_id:
"""
pdf = ds_to_pdf(ds, user_id)
data = []
for index, row in pdf.iterrows():
data.append(
dict(Task=row["stresser_sub"],
Start=row["start_time"],
Finish=row["end_time"],
Resource=row["stresser_main"]))
fig = ff.create_gantt(data,
index_col='Resource',
title='Stressers, Main & Sub Categories',
show_colorbar=True,
bar_width=0.8,
showgrid_x=True,
showgrid_y=True)
fig['layout']['yaxis'].update({"showticklabels": False})
iplot(fig, filename='gantt-hours-minutes')
def plot_bar(ds, x_axis_column="stresser_main"):
"""
Args:
ds:
user_id:
x_axis_column:
"""
pdf = ds_to_pdf(ds)
grouped_pdf = pdf.groupby(["user", x_axis_column],
as_index=False).agg('mean')
user_ids = pdf.groupby("user", as_index=False).last()
data = []
for index, row in user_ids.iterrows():
sub = grouped_pdf.loc[grouped_pdf['user'] == row["user"]]
sub.sort_values(x_axis_column)
data.append(
go.Bar({
'y': sub["density"],
'x': sub[x_axis_column],
'name': row["user"]
}))
layout = go.Layout(
title="All Participants' Stress Levels By Each Stressors",
yaxis=dict(title='Average Stress Density'))
fig = go.Figure(data=data, layout=layout)
iplot(fig, filename='basic-line')
def plot_sankey(ds,
user_id,
cat_cols=["stresser_main", "stresser_sub"],
value_cols='density',
title="Stressers' Sankey Diagram"):
"""
Args:
ds:
user_id:
cat_cols:
value_cols:
title:
"""
pdf = ds_to_pdf(ds, user_id)
labelList = []
for catCol in cat_cols:
labelListTemp = list(set(pdf[catCol].values))
labelList = labelList + labelListTemp
# remove duplicates from labelList
labelList = list(dict.fromkeys(labelList))
# transform df into a source-target pair
for i in range(len(cat_cols) - 1):
if i == 0:
sourceTargetDf = pdf[[cat_cols[i], cat_cols[i + 1], value_cols]]
sourceTargetDf.columns = ['source', 'target', 'density']
else:
tempDf = pdf[[cat_cols[i], cat_cols[i + 1], value_cols]]
tempDf.columns = ['source', 'target', 'density']
sourceTargetDf = pd.concat([sourceTargetDf, tempDf])
sourceTargetDf = sourceTargetDf.groupby(['source', 'target']).agg({
'density':
'mean'
}).reset_index()
# add index for source-target pair
sourceTargetDf['sourceID'] = sourceTargetDf['source'].apply(
lambda x: labelList.index(x))
sourceTargetDf['targetID'] = sourceTargetDf['target'].apply(
lambda x: labelList.index(x))
# creating the sankey diagram
data = dict(type='sankey',
node=dict(pad=15,
thickness=20,
line=dict(color="black", width=0.5),
label=labelList),
link=dict(source=sourceTargetDf['sourceID'],
target=sourceTargetDf['targetID'],
value=sourceTargetDf['density']))
layout = dict(title=title, font=dict(size=10))
fig = dict(data=[data], layout=layout)
iplot(fig, validate=False)
def plot_timeseries(ds,
user_id: str = None,
x_axis_column: str = "timestamp",
y_axis_column: list = "all",
graph_title: str = "Graph"):
"""
line plot of timeseries data
Args:
ds:
user_id (str): uuid of a user
x_axis_column (str): timestamp or localtime are acceptable values only
y_axis_column (list): set this to "all" if you want to plot all columns
graph_title (str): title of the graph
"""
pdf = ds_to_pdf(ds, user_id)
user_ids = list(pdf.user.unique())
subplot_titles = [
"Participant ID: {}".format(x.upper()) for x in list(pdf.user.unique())
]
if x_axis_column not in ["timestamp", "localtime"]:
raise Exception(" X axis can only be timestamp or localtime column.")
if y_axis_column == "all":
y_axis = sorted(
list(
set(pdf.columns.to_list()) -
set(["timestamp", "user", "version", "localtime"])))
elif isinstance(y_axis_column, list):
y_axis = y_axis_column
elif isinstance(y_axis_column, str):
y_axis = [y_axis_column]
fig = make_subplots(rows=len(user_ids),
cols=1,
subplot_titles=subplot_titles)
row_id = 1
for sid in user_ids:
if user_ids[-1] == sid:
lagend = True
else:
lagend = False
for y in y_axis:
fig.add_trace(go.Scatter(x=pdf[x_axis_column][pdf.user == sid],
y=pdf[y][pdf.user == sid],
showlegend=lagend,
name=y),
row=row_id,
col=1)
fig.update_xaxes(title_text="Timestamp", row=row_id, col=1)
row_id += 1
height = 500 * len(user_ids)
fig.update_layout(height=height, width=900, title_text=graph_title)
fig.show()
def plot_comparison(ds,
x_axis_column="stresser_main",
usr_id=None,
compare_with="all"):
"""
Args:
ds:
x_axis_column:
usr_id:
compare_with:
"""
pdf = ds_to_pdf(ds)
data = []
if usr_id:
usr_data = pdf.loc[pdf['user'] == str(usr_id)]
if compare_with == "all" or compare_with is None:
compare_with_data = pdf.loc[pdf['user'] != str(usr_id)]
else:
compare_with_data = pdf.loc[pdf['user'] == str(compare_with)]
grouped_user_pdf = usr_data.groupby([x_axis_column],
as_index=False).agg('mean')
grouped_compare_with_pdf = compare_with_data.groupby(
[x_axis_column], as_index=False).agg('mean')
data.append(
go.Bar({
'y': grouped_user_pdf["density"],
'x': grouped_user_pdf[x_axis_column],
'name': usr_id
}))
if compare_with == "all":
compare_with = "All Participants"
data.append(
go.Bar({
'y': grouped_compare_with_pdf["density"],
'x': grouped_compare_with_pdf[x_axis_column],
'name': compare_with
}))
layout = go.Layout(
title="Comparison of Stress Levels Amongst Participants",
yaxis=dict(title='Average Stress Density'))
fig = go.Figure(data=data, layout=layout)
iplot(fig, filename='basic-line')
else:
raise Exception("usr_id cannot be None/Blank.")
def plot_pie(ds, user_id, group_by_column="stresser_main"):
"""
Args:
ds:
user_id:
group_by_column:
"""
pdf = ds_to_pdf(ds, user_id)
pdf = pdf.groupby(str(group_by_column), as_index=False).agg('count')
labels = []
values = []
for index, row in pdf.iterrows():
labels.append(row["stresser_main"])
values.append(row["density"])
trace = go.Pie(labels=labels, values=values)
iplot([trace], filename='stresser_pie_chart')
def plot_gps_clusters(ds, user_id: str = None, zoom=10):
"""
Plots GPS coordinates
Args:
ds (DataStream): datastream object
user_id (str): uuid of a user
zoom: min 0 and max 100, zoom map
"""
pdf = ds_to_pdf(ds, user_id)
center = None
for index, row in pdf.iterrows():
if center is None:
center = [row["latitude"], row["longitude"]]
m = folium.Map(location=center, zoom_start=zoom)
folium.Marker(location=(row["latitude"], row["longitude"])).add_to(m)
return m
def plot_hist(ds, user_id: str, x_axis_column=None):
"""
histogram plot of timeseries data
Args:
ds (DataStream):
user_id (str): uuid of a user
x_axis_column (str): x axis column of the plot
"""
pdf = ds_to_pdf(ds, user_id)
cf.set_config_file(offline=True, world_readable=True, theme='ggplot')
init_notebook_mode(connected=True)
pdf = _remove_cols(pdf)
if x_axis_column:
data = [go.Histogram(x=pdf[str(x_axis_column)])]
iplot(data, filename='basic histogram')
else:
pdf.iplot(kind='histogram', filename='basic histogram')
def plot_timeseries(ds: DataStream, user_id: str, y_axis_column: str = None):
"""
line plot of timeseries data
Args:
ds (DataStream):
user_id (str): uuid of a user
y_axis_column (str): x axis column is hard coded as timestamp column. only y-axis can be passed as a param
"""
pdf = ds_to_pdf(ds, user_id)
cf.set_config_file(offline=True, world_readable=True, theme='ggplot')
init_notebook_mode(connected=True)
ts = pdf['timestamp']
pdf = _remove_cols(pdf)
if y_axis_column:
data = [go.Scatter(x=ts, y=pdf[str(y_axis_column)])]
iplot(data, filename='time-series-plot')
else:
iplot([{
'x': ts,
'y': pdf[col],
'name': col
} for col in pdf.columns],
filename='time-series-plot')