# save_dist = df_dist
# else:
# save_dist = pd.concat([save_dist,df_dist],axis=1, join='inner')
# outputFile = "output/Dist_tables/%s_Dist.csv" %file_name
# os.makedirs(os.path.dirname(outputFile), exist_ok=True)
# if os.path.exists(outputFile) == False:
# with open(outputFile, 'w') as f:
# save_dist.to_csv(f)
# elif os.path.exists(outputFile) == True:
# with open(outputFile, 'a') as f:
# save_dist.to_csv(f)
# ##################### END ########################
for c in range(0, len(col)):
func.dist_Plot(combined_df, col[c], ctrl_var)
elif ctrl_var in col:
list_value = list(Counter(combined_df[ctrl_var]).keys())
if len(list_value) < 6:
for i_value in list_value:
ctrl_combined_df = combined_df[combined_df[ctrl_var] ==
i_value]
for c in range(0, len(col)):
if ctrl_var != col[c]:
func.dist_Plot(ctrl_combined_df, col[c],
str(ctrl_var + '_' + str(i_value)))
else:
logger.error(
"Sorry, control variable has too many different values! Please choose categorical variable as control"
)
if input['check_missing'] == True:
func.check_missing(df, col, year)
if input['data_description'] == True:
func.data_describe(df, col, year)
### For three plots ###
loop = input['age_range']
for i in loop:
df_avg = func.groupAgeRange(df_vektis, i, 0)
if input['correlation_matrix'] == True:
func.corr_Matrix(df_avg, i, year)
if input['pie_chart'] == True:
func.pie_Chart(df_avg, i, year)
if input['distribution_plot'] == True:
func.dist_Plot(df_avg, 'SUM', i, year)
### Only for the Stack plot ###
if input['stacked_area'] == True:
loop = list(range(0, 90, 1))
df_stack = pd.DataFrame()
for i in loop:
df_avg = func.groupAgeRange(df_vektis, i, df_stack)
df_stack[i] = df_avg.mean(axis=0, skipna=True)
df_stack_trans = df_stack.transpose()
df_stack_trans = func.merge(df_stack_trans)
func.stacked_Plot(df_stack_trans, year)
print('Stacked Area plot is done')
### Separate features to numerical and categorical ###
numFea = []
catFea = []
for c in col:
if len(Counter(df[c].dropna())) > 20:
numFea.append(c)
else:
catFea.append(c)
### Function for distribution plot ###
if inputYAML['distribution_plot'] == True:
if inputYAML['distribution_feature'] == 'ALL':
for f in numFea:
try:
func.dist_Plot(df[numFea], f, file_name)
except:
if f not in catFea:
logger.error(f, " -- Data type does not support numerical distribution plot")
for f in catFea:
try:
func.cate_Dist(df[catFea], f, file_name)
except:
if f not in numFea:
logger.error(f, " -- Data type does not support categorical distribution plot")
else:
for f in inputYAML['distribution_feature']:
if f in numFea:
try: