groupby = 'Reoperation'
categorical = []
for element in columns:
if element not in remove_cols:
categorical.append(element)
mytable = TableOne(data=df,
columns=list_columns,
categorical=categorical,
groupby=groupby,
pval=True,
smd=True,
htest_name=True) #,min_max=remove_cols)
#data, columns, categorical, groupby, nonnormal, pval = True, smd=True,htest_name=True
mytable.to_csv("_table.csv")
############################################################################
##merge
df = pd.read_csv("op_first_table_table.csv")
df1 = pd.read_csv("op_reop_table_table.csv")
df = df.rename(
columns={
'Unnamed: 0': 'values',
'Unnamed: 1': 'type',
'Missing': 'Missing_first',
'Overall': 'first'
})
df1 = df1.rename(
# In[276]:
## 非正态分布列名列表
print(mytable)
"""
Display the table using the tabulate method. The tablefmt argument allows the table to be displayed in multiple formats,
including “github”, “grid”, “fancy_grid”, “rst”, “html”, and “latex”.
"""
# In[261]:
mytable.to_csv('pythontable1.csv')
# In[262]:
help(TableOne)
# In[ ]:
#Sort the data in the right samples
data1.sort_values(by=['year'], inplace=True)
df_1_t = data1.iloc[:94, :] #all the samples of 2019
df_2 = data1.iloc[95:, :] #all the samples of 2020
#drop_indices = np.random.choice(df_2_t.index, 3, replace=False) # make the sample sizes of 2020 and 2019 same length by deleting three random entries
#df_2=df_2_t.drop(drop_indices)
df_1 = df_1_t.drop(57) #drop the nan data sample
df_1.sort_values(by=['living'], inplace=True) #sort by living situation
df_2.sort_values(by=['living'], inplace=True) #sort by living situation
#make table 1
columns = ['year', 'gender', 'bmi', 'living']
mytable = TableOne(data1, columns=columns, pval=False)
print(mytable.tabulate(tablefmt="fancy_grid"))
mytable.to_csv('mytable.csv')
#split the data samples in living with their parents and moved out for 2020 and calculate means
grouped1 = df_1.groupby(df_1.living)
Moved_out_2019 = grouped1.get_group("Moved_out")
mean1 = Moved_out_2019["attitu_2"].mean()
Parents_2019 = grouped1.get_group("Living_with_parents")
mean2 = Parents_2019["attitu_2"].mean()
#split the data samples in living with their parents and moved out for 2019 and calculate means
grouped2 = df_2.groupby(df_2.living)
Moved_out_2020 = grouped2.get_group("Moved_out")
mean3 = Moved_out_2020["attitu_2"].mean()
Parents_2020 = grouped2.get_group("Living_with_parents")
mean4 = Parents_2020["attitu_2"].mean()
#plot in a figure the perception values
MYCATEGORICAL = set()
for name, mytype in zip(data.dtypes.index, data.dtypes.values):
if mytype == object:
MYCATEGORICAL.add(name)
for c in categorical:
MYCATEGORICAL.add(c)
MYCATEGORICAL = list(MYCATEGORICAL) if len(MYCATEGORICAL) > 0 else None
# create grouped_table with p values
# API: https://tableone.readthedocs.io/en/latest/tableone.html
# Example: https://github.com/tompollard/tableone/blob/master/tableone.ipynb
grouped_table = TableOne(data,
categorical=MYCATEGORICAL,
groupby=group,
nonnormal=nonnormal,
label_suffix=True,
pval=pval)
print(grouped_table)
if outfile:
if outFMT == 'csv':
grouped_table.to_csv(outfile)
elif outFMT == 'latex':
grouped_table.to_latex(outfile)
elif outFMT == 'html':
grouped_table.to_html(outfile)
sys.stdout.flush()
sys.stdout.close()
sys.stderr.flush()
sys.stderr.close()