def make_race(drug_name=base_drug):
"""
Ceates bar plot relating the race of the victims to the specific drug
and returns bar plot race
Parameters
----------
drug_name : Chosen from from the dropdown menu
Returns
-------
Plots: race
Examples
--------
>>> ake_demographics(drug_name = 'Heroin')
"""
if drug_name == 'Everything':
query = pivoted_data
else:
query = pivoted_data.query(drug_name + ' == 1')
race = alt.Chart(query).mark_bar().encode(
x=alt.X("Race:N", title="Race", axis=alt.AxisConfig(labelAngle=-45)),
y='count()',
color=alt.Color('Race:N',
scale=alt.Scale(scheme='viridis'),
legend=None)).properties(
title='Race distribution for ' + drug_name,
width=400,
height=180)
return race
def build_chart(self):
"""
Altair Chart constructed using the selected data from SimpleSpectrum and SimpleSpectralLines objects, along with the set chart parameters
Returns
-------
The Altair Chart
"""
layer_list = []
for spectrum in self.spectrum_dict.values():
layer = spectrum.build_chart(
wavelength_axis_label=self._get_wavelength_title(),
flux_axis_label=self._get_flux_title(),
wavelength_unit=self.wavelength_unit)
if layer is not None:
layer_list.append(layer)
if self.lines is not None:
_, wavelength_min, wavelength_max = self.get_all_spectrum_chart_data(
)
self.lines.set_wavelength_limits(wavelength_min=wavelength_min,
wavelength_max=wavelength_max)
layer_list.extend(
self.lines.build_chart(
wavelength_axis_label=self._get_wavelength_title(),
wavelength_unit=self.wavelength_unit))
base_chart = alt.layer(*layer_list, ).configure(
axis=alt.AxisConfig(grid=self._show_grid),
legend=alt.LegendConfig(disable=not (self._show_legend)),
).properties(**self._chart_properties).interactive()
return base_chart
def make_race(drug_name="Amphet"):
query = data.query(drug_name + ' == 1')
race = alt.Chart(query).mark_bar().encode(
x=alt.X("Race:N", title="Race", axis=alt.AxisConfig(labelAngle=45)),
y='count()',
color=alt.Color('Race:N',
scale=alt.Scale(scheme='viridis'))).properties(
title='Race distribution for ' + drug_name,
width=400,
height=180)
return race
def concat(plot_1, plot_2, dash_name):
plot = alt.vconcat(plot_1, plot_2, spacing=60).configure(
legend=alt.LegendConfig(labelFontSize=16,
titleFontSize=16,
symbolSize=100,
labelFont='Lato'),
axis=alt.AxisConfig(labelFontSize=16,
tickSize=16,
labels=True,
titleFontSize=16),
header=alt.HeaderConfig(titleFontSize=20,
labelFontSize=20)).configure_title(fontSize=24)
plot.save(f"app/templates/plot/{dash_name}.html")
def make_demographics(drug_name=base_drug):
"""
Contains the mds theme and creates two bar plots relating the age and the gender of the victims to the specific drug
and returns two plots age and gender
Parameters
----------
drug_name : Chosen from from the dropdown menu
Returns
-------
Plots: age | gender
Examples
--------
>>> make_demographics(drug_name = 'Heroin')
"""
def mds_special():
font = "Arial"
axisColor = "#000000"
gridColor = "#DEDDDD"
return {
"config": {
"title": {
"fontSize": 24,
"font": font,
"anchor": "start", # equivalent of left-aligned.
"fontColor": "#000000"
},
'view': {
"height": 300,
"width": 400
},
"axisX": {
"domain": True,
#"domainColor": axisColor,
"gridColor": gridColor,
"domainWidth": 1,
"grid": False,
"labelFont": font,
"labelFontSize": 12,
"labelAngle": 0,
"tickColor": axisColor,
"tickSize":
5, # default, including it just to show you can change it
"titleFont": font,
"titleFontSize": 16,
"titlePadding":
10, # guessing, not specified in styleguide
"title": "X Axis Title (units)",
},
"axisY": {
"domain": False,
"grid": True,
"gridColor": gridColor,
"gridWidth": 1,
"labelFont": font,
"labelFontSize": 14,
"labelAngle": 0,
#"ticks": False, # even if you don't have a "domain" you need to turn these off.
"titleFont": font,
"titleFontSize": 16,
"titlePadding":
10, # guessing, not specified in styleguide
"title": "Y Axis Title (units)",
# titles are by default vertical left of axis so we need to hack this
#"titleAngle": 0, # horizontal
#"titleY": -10, # move it up
#"titleX": 18, # move it to the right so it aligns with the labels
},
}
}
# Register the custom theme under a chosen name
alt.themes.register('mds_special', mds_special)
# Enable the newly registered theme
alt.themes.enable('mds_special')
# Creat plots
sub_data = pivoted_data.query("Sex == 'Male' | Sex == 'Female'")
if drug_name == 'Everything':
query = sub_data
else:
query = sub_data.query(drug_name + ' == 1')
chart = alt.Chart(query)
age = chart.mark_bar(color="#3f7d4e").encode(
x=alt.X("Age:Q",
title="Age",
bin=alt.Bin(maxbins=10),
axis=alt.AxisConfig(labelAngle=-45)),
y='count()').properties(title='Age distribution for ' + drug_name,
width=290,
height=200)
gender = chart.mark_bar().encode(
x=alt.X("Sex:N", title="Sex", axis=alt.AxisConfig(labelAngle=-45)),
y='count()',
color=alt.Color('Sex:N',
scale=alt.Scale(scheme='viridis'),
legend=None)).properties(
title='Gender distribution for ' + drug_name,
width=190,
height=200)
return (age | gender)
def main(input, output):
X_train = pd.read_csv(input + "/" + "X_train.csv")
y_train = pd.read_csv(input + "/" + "y_train.csv")
X_test = pd.read_csv(input + "/" + "X_test.csv")
y_test = pd.read_csv(input + "/" + "y_test.csv")
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)
df = pd.DataFrame(
columns=['n_features_to_select', 'Train_error', 'Test_error'])
for n in range(1, 11):
lr = LinearRegression()
rfe = RFE(estimator=lr, n_features_to_select=n)
rfe.fit(X_train, y_train)
X_train_sel = X_train_scaled[:, rfe.support_]
X_test_sel = X_test_scaled[:, rfe.support_]
errors = fit_and_report(lr,
X_train_sel,
y_train,
X_test_sel,
y_test,
mode='regression')
df = df.append(
{
'n_features_to_select': n,
'Train_error': errors[0],
'Test_error': errors[1]
},
ignore_index=True)
df = pd.melt(df,
id_vars=['n_features_to_select'],
value_vars=['Train_error', 'Test_error'],
var_name='error_type',
value_name='value')
feature_plot = alt.Chart(df).mark_line().encode(
x=alt.X('n_features_to_select:N',
title='Number of features to select',
axis=alt.AxisConfig(labelAngle=0)),
y='value:Q',
color=alt.Color('error_type',
sort=['Train error'])).configure_scale(round=True)
feature_plot.configure_header(titleFontSize=80, labelFontSize=80)
feature_plot.configure().properties(
title="The relationship between MSE and number of features",
width=800,
height=400).save(output + "/ranked_features.png")
lr = LinearRegression()
rfe = RFE(estimator=lr, n_features_to_select=6)
rfe.fit(X_train, y_train)
X_train_sel = X_train_scaled[:, rfe.support_]
X_test_sel = X_test_scaled[:, rfe.support_]
lr.fit(X_train_sel, y_train)
errors = fit_and_report(lr,
X_train_sel,
y_train,
X_test_sel,
y_test,
mode='regression')
print(errors)
relevant_features_bool = rfe.support_
relevant_features_list = pd.DataFrame(
list(X_train.iloc[:, relevant_features_bool].columns))
relevant_features_list
relevant_features_list['weights'] = lr.coef_[0]
relevant_features_list = relevant_features_list.rename(
columns={0: "features"})
# test
assert len(
relevant_features_list) == 6, 'The dimension of y_pred_df is wrong'
print(relevant_features_list)
feature_weight_plot = alt.Chart(relevant_features_list).mark_bar().encode(
alt.Y('features:N',
sort=alt.EncodingSortField(field="features",
op="count",
order='ascending')),
alt.X('weights:Q'))
feature_weight_plot.configure_header(titleFontSize=80, labelFontSize=80)
feature_weight_plot.configure().properties(
title="The feature weights", width=800,
height=400).save(output + "/feature_weight_plot.png")
y_pred_df = pd.DataFrame(lr.predict(X_test_sel))
y_pred_df = y_pred_df.rename(columns={0: "predicted"})
y_true_df = y_test.rename(columns={'quality': "actual"})
result_df = pd.concat([y_pred_df, y_true_df], axis=1)
plot_result = alt.Chart(result_df).mark_boxplot().encode(
alt.X('actual:O',
scale=alt.Scale(zero=False),
axis=alt.AxisConfig(labelAngle=0)),
alt.Y('predicted', scale=alt.Scale(zero=False)))
plot_result.configure_header(titleFontSize=80, labelFontSize=80)
plot_result.configure(numberFormat="0.4f").properties(
title="The prediction result", width=800,
height=400).save(output + "/prediction_result.png")
print('This is the end of analysis, Hooray!!!')