def histograms(days):
data = [
Histogram(
x=[
day.overtime.total_seconds() / 3600 for day in days
if day.overtime > datetime.timedelta()
],
name="overtime",
xbins={"size": 1},
opacity=0.75,
),
Histogram(
x=[
day.minus_hours.total_seconds() / 3600 for day in days
if day.minus_hours < datetime.timedelta()
],
name="minus hours",
xbins={"size": 1},
opacity=0.75,
),
]
layout = Layout(
title="Overtime / Minus hours",
xaxis={"title": "Overtime / Minus hours"},
yaxis={"title": "Occurrences (in days)"},
barmode="overlay",
)
return dict(data=data, layout=layout)
def _histogram(self, names, x):
concatenated_numeric_values = self._get_numeric_values(
self._concatenated_data_frame, x)
start = min(concatenated_numeric_values)
end = max(concatenated_numeric_values)
size = 1 if str(self._concatenated_data_frame[x].dtype) == 'category' \
else (max(concatenated_numeric_values) - min(concatenated_numeric_values)) / 10.0
fig = tools.make_subplots(rows=1, cols=len(names), print_grid=False)
histogram_index = 1
for name in names:
df = self._dataframes[name]
numeric_values = self._get_numeric_values(df, x)
text = df[x].cat.categories if str(
df[x].dtype) == 'category' else None
histogram = Histogram(
name=name,
x=numeric_values,
xbins=dict(
start=start,
end=end,
size=size,
),
text=text,
)
fig.append_trace(histogram, 1, histogram_index)
fig.layout['xaxis' + str(histogram_index)].title = x
fig.layout['xaxis' + str(histogram_index)].range = [start, end]
fig.layout['yaxis' + str(histogram_index)].title = 'count'
histogram_index += 1
fig.layout.width = min(500 * len(names), 1200)
fig.layout.height = 500
iplot(fig)
def index():
## For genre analysis
# extract data needed for visuals
genre_counts = df.groupby('genre').count()['message']
genre_names = list(genre_counts.index)
pos_ratios = list((df[df.columns[4:]] > 0).mean(axis=1))
cat_names = list(df.columns[4:])
message_lengths = df.message.apply(
lambda text: len(TextBlob(text).tokens)).values
# create visuals
graphs = [{
'data': [Bar(
x=genre_names,
y=genre_counts,
)],
'layout': {
'title': 'Distribution of Message Genres',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Genre"
}
}
}, {
'data': [Bar(x=cat_names, y=pos_ratios)],
'layout': {
'title': 'Distribution of Non-Zero labels in Each Category',
'yaxis': {
'title': "Ratio of Positive Instances"
},
'xaxis': {
'title': "Category Name"
}
}
}, {
'data': [
Histogram(x=message_lengths,
xbins=dict(start=np.min(message_lengths),
size=0.8,
end=np.max(message_lengths)))
],
'layout': {
'title': 'Distribution of Message Length',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Message Length"
}
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def index():
# extract data needed for visuals
# TODO: Below is an example - modify to extract data for your own visuals
genre_counts = df.groupby('genre').count()['message']
genre_names = list(genre_counts.index)
# create visuals
# TODO: Below is an example - modify to create your own visuals
category_map = df.iloc[:, 4:].corr().values
category_names = list(df.iloc[:, 4:].columns)
category_names = df.iloc[:, 4:].columns
category_boolean = (df.iloc[:, 4:] != 0).sum().values
graphs = [{
'data': [Bar(x=genre_names, y=genre_counts)],
'layout': {
'title': 'Distribution of Message Genres',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Genre"
}
}
}, {
'data': [Histogram(x=df.length_under_600, )],
'layout': {
'title': 'Histogram of Messages Length below 600 character',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Length of Message"
}
}
}, {
'data': [Bar(x=category_names, y=category_boolean)],
'layout': {
'title': 'Distribution of Message Categories',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Category",
'tickangle': 35
}
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def index():
# extract data needed for visuals
# TODO: Below is an example - modify to extract data for your own visuals
counts = df.iloc[:, 4:].apply(lambda x: x.value_counts()).T
counts.sort_values(by=1, ascending=False, inplace=True)
categories = pd.Series(counts.index).str.replace('_', ' ').str.title()
row_sums = df.iloc[:, 4:].sum(axis='columns')
# create visuals
# TODO: Below is an example - modify to create your own visuals
colors = [
plotly.colors.DEFAULT_PLOTLY_COLORS[i % len(
plotly.colors.DEFAULT_PLOTLY_COLORS)] for i in range(len(counts))
]
graphs = [{
'data': [Bar(
x=categories,
y=counts[1],
marker=dict(color=colors),
)],
'layout': {
'title': 'Distribution of Message Categories',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Category",
'tickangle': -90
},
'margin': {
'b': 160
}
}
}, {
'data': [Histogram(x=row_sums)],
'layout': {
'title': 'Distribution of Counts of Multiple Categories',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Number of categories in message"
}
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def index():
# extract data needed for visuals
genre_counts = df.groupby('genre').count()['message']
genre_names = list(genre_counts.index)
cat_frequency = df[df.columns[4:]].sum().sort_values(ascending=False)[:10]
cats = cat_frequency.index
msg_len = df.message.apply(lambda x: len(x.split()))
msg_len = msg_len[~((msg_len - msg_len.mean()).abs() > 3 * msg_len.std())]
# create visuals
graphs = [{
'data': [Bar(x=cats, y=cat_frequency, marker=dict(color='#031c57'))],
'layout': {
'title': 'Top 10 Message Categories',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Category"
}
}
}, {
'data': [Histogram(x=msg_len, marker=dict(color='#031c57'))],
'layout': {
'title': 'Distribution of Message Lengths',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Words/Message"
}
}
}, {
'data':
[Bar(x=genre_names, y=genre_counts, marker=dict(color='#031c57'))],
'layout': {
'title': 'Message Genres',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Genre"
}
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def index():
# extract data needed for visuals
genre_counts = df.groupby('genre').count()['message']
genre_names = list(genre_counts.index)
categories = df.iloc[:, 4:]
cat_names = list(categories)
cat_counts = [df[cat_name].sum() for cat_name in cat_names]
categories['sum'] = categories.sum(axis=1)
counts = categories.groupby('sum').count()['related']
names = list(counts.index)
# create visuals
#custom plotly graphs
graphs = [{
'data': [Histogram(
y=counts,
marker=dict(color='#4CB391'),
)],
'layout': {
'title': 'Distribution of Messages in several categories',
'yaxis': {
'title': "Number of messages"
},
'xaxis': {
'title': "Number of included categories"
},
}
}, {
'data': [Bar(x=cat_names, y=cat_counts, marker=dict(color='#4CB391'))],
'layout': {
'title': 'Distribution of Message Categories',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Category"
}
}
}, {
'data': [Pie(labels=genre_names, values=genre_counts)],
'layout': {
'title': 'Distribution of Message Genres',
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def index():
# extract data needed for visuals
genre_counts = df.groupby('genre').count()['message']
genre_names = list(genre_counts.index)
#proportion of categories
category_counts = df[df.columns[4:]].sum() / len(df)
category_counts.sort_values(ascending=False, inplace=True)
category_names = list(category_counts.index)
df['avg_categories'] = df[df.columns[4:]].sum(axis=1)
# create visuals
graphs = [{
'data': [Bar(x=genre_names, y=genre_counts)],
'layout': {
'title': 'Distribution of Message Genres',
'yaxis': {
'title': "Count"
},
'xaxis': {
'title': "Genre"
}
}
}, {
'data': [Bar(x=category_names, y=category_counts)],
'layout': {
'title': 'Distribution of Categories',
'yaxis': {
'title': "Proportion"
},
'xaxis': {
'title': "Categories"
}
}
}, {
'data': [Histogram(x=df['avg_categories'], )],
'layout': {
'title': 'Number of Categories per message',
'yaxis': {
'title': "Number of messages"
},
'xaxis': {
'title': "Categories"
}
}
}]
# encode plotly graphs in JSON
ids = ["graph-{}".format(i) for i, _ in enumerate(graphs)]
graphJSON = json.dumps(graphs, cls=plotly.utils.PlotlyJSONEncoder)
# render web page with plotly graphs
return render_template('master.html', ids=ids, graphJSON=graphJSON)
def showPlotlyhistogram(self):
# Plotly Offline
data = fd.data[fd.logcolumn]
offline.plot({
"data": [
Histogram(x=data,
xbins=dict(start=int(fd.fmin),
end=int(fd.fmax),
size=fd.binwidth))
],
"layout":
Layout(title="Log10(D) histogram for " + self.cellid)
})