def loan_mapping(map_rendered=False):
mform = map_form(request.form)
if request.method == 'POST' and mform.validate(
): #if user is posting, get form data and store it
map_rendered = True
# compute desired group-by/agg operation
data = LCH.compute_group_avgs(LD,
mform.data['col_name'],
mform.data['grouping_var'],
mform.data['agg_fun'],
state_fips_dict=state_fips_dict,
min_counts=50)
data.name = mform.data['col_name']
if mform.data['agg_fun'] == 'count':
data.name = 'counts'
# paint base map by county
pal = LCH.paint_map(data,
app.base_map,
app.county_paths,
fips_to_zip,
color='cube',
name_legend_map=name_legend_map,
agg_fun=mform.data['agg_fun'])
# save colorbar for map as a png
plt.savefig(fig_dir + 'map_cbar.png', dpi=500, format='png')
plt.close()
return render_template('loan_mapping.html',
map_form=mform,
svg=Markup(str(app.base_map)),
rnum=np.random.randint(0, 100000),
map_rendered=map_rendered)
def loan_mapping(map_rendered=False):
mform = map_form(request.form)
if request.method == 'POST' and mform.validate(): #if user is posting, get form data and store it
map_rendered = True
# compute desired group-by/agg operation
data = LCH.compute_group_avgs(LD, mform.data['col_name'],
mform.data['grouping_var'],
mform.data['agg_fun'],
state_fips_dict=state_fips_dict,
min_counts=50)
data.name = mform.data['col_name']
if mform.data['agg_fun'] == 'count':
data.name = 'counts'
# paint base map by county
pal = LCH.paint_map(data, app.base_map, app.county_paths, fips_to_zip,
color='cube', name_legend_map=name_legend_map,
agg_fun=mform.data['agg_fun'])
# save colorbar for map as a png
plt.savefig(fig_dir + 'map_cbar.png', dpi=500, format='png')
plt.close()
return render_template('loan_mapping.html', map_form=mform, svg=Markup(str(app.base_map)),
rnum=np.random.randint(0,100000), map_rendered=map_rendered)
data_dir = os.path.join(base_dir,'static/data/')
fig_dir = os.path.join(base_dir,'static/images/')
movie_dir = os.path.join(base_dir,'static/movies/')
data_name = 'all_loans_proc'
LD = pd.read_csv(data_dir + data_name, parse_dates=['issue_d'])
fips_data = LCL.load_location_data(data_dir, group_by='fips')
zip3_data = LCL.load_location_data(data_dir, group_by='zip3')
fips_to_zip = LCL.make_fips_to_zip_dict(data_dir, group_by='zip')
#%% make a k-tree for doing nearest neighbor imputation of missing data
base_map = LCL.load_base_map(fig_dir + 'USA_Counties_text.svg', ax_xml=True)
(county_paths,state_paths) = LCL.get_map_paths(base_map,fips_to_zip)
title_path = base_map.findAll('text')[0]
map_coords = LCH.extract_fips_coords(county_paths)
ktree = KDTree(map_coords.values) #make nearest neighbor tree
#%% make sequence of decision trees and build a movie
X = LD[['longitude','latitude']]
y = LD['ROI'] #plot average return by area, not portfolio return
max_levels = 16
min_samples_leaf = 50
pred_arr = np.zeros((len(fips_data),max_levels))
for n in xrange(max_levels):
clf = tree.DecisionTreeRegressor(max_depth=n+1, min_samples_leaf=min_samples_leaf, random_state=0)
clf.fit(X, y)
pred_arr[:,n] = clf.predict(fips_data[['longitude','latitude']].values)
#%% generate pngs for reg-tree at each depth value
all_hazards[term] = np.zeros((len(keep_grades),term+1)) #initialize matrix of hazard functions
for gidx,grade in enumerate(keep_grades): #fit model for each grade
grade_data = cur_data.grade == grade
naf.fit(lifetimes[grade_data],event_observed=is_observed[grade_data],label=grade,timeline=np.arange(term+1))
all_hazards[term][gidx,:] = naf.smoothed_hazard_(term_bandwidths[idx]).squeeze()
#%%
terms = LD.term.unique() #set of unique loan terms
for term in terms: #for each possible loan term
#get relevant set of loans
cur_loans = LD.term == term
cur_LD = LD[cur_loans]
(NAR, net_returns, p_csum) = LCH.get_NARs(cur_LD, term)
LD.ix[cur_loans,'ROI'] = NAR #measured performance of each loan
LD.ix[cur_loans,'net_returns'] = net_returns #principal weighted avg monthly returns
LD.ix[cur_loans,'prnc_weight'] = p_csum #principal weighted avg monthly returns
LD.ix[cur_loans,'default_prob'] = LD.ix[cur_loans,'is_observed'].astype(float) #principal weighted avg monthly returns
(exp_NAR, tot_default_prob, exp_num_pymnts, exp_net_returns, exp_csum) = \
LCH.get_expected_NARs(cur_LD, term, all_hazards[term])
LD.ix[cur_loans,'ROI'] = exp_NAR
LD.ix[cur_loans,'default_prob'] = tot_default_prob
LD.ix[cur_loans,'exp_num_pymnts'] = exp_num_pymnts
LD.ix[cur_loans,'net_returns'] = exp_net_returns
LD.ix[cur_loans,'prnc_weight'] = exp_csum
LD.ix[cur_loans, 'best_NAR'] = LCH.get_best_returns(cur_LD, term)
grade_data = cur_data.grade == grade
naf.fit(lifetimes[grade_data],
event_observed=is_observed[grade_data],
label=grade,
timeline=np.arange(term + 1))
all_hazards[term][gidx, :] = naf.smoothed_hazard_(
term_bandwidths[idx]).squeeze()
#%%
terms = LD.term.unique() #set of unique loan terms
for term in terms: #for each possible loan term
#get relevant set of loans
cur_loans = LD.term == term
cur_LD = LD[cur_loans]
(NAR, net_returns, p_csum) = LCH.get_NARs(cur_LD, term)
LD.ix[cur_loans, 'ROI'] = NAR #measured performance of each loan
LD.ix[cur_loans,
'net_returns'] = net_returns #principal weighted avg monthly returns
LD.ix[cur_loans,
'prnc_weight'] = p_csum #principal weighted avg monthly returns
LD.ix[cur_loans, 'default_prob'] = LD.ix[cur_loans, 'is_observed'].astype(
float) #principal weighted avg monthly returns
(exp_NAR, tot_default_prob, exp_num_pymnts, exp_net_returns, exp_csum) = \
LCH.get_expected_NARs(cur_LD, term, all_hazards[term])
LD.ix[cur_loans, 'ROI'] = exp_NAR
LD.ix[cur_loans, 'default_prob'] = tot_default_prob
LD.ix[cur_loans, 'exp_num_pymnts'] = exp_num_pymnts
LD.ix[cur_loans, 'net_returns'] = exp_net_returns
LD.ix[cur_loans, 'prnc_weight'] = exp_csum
