tableone
tableone copied to clipboard
Published
20 hours ago •
tompollard
tompollard
improve output format
The current approach relies on pandas to output the table to tex, csv, etc. Add custom approach to improve the output quality.
Consider tabulate:
from tableone import TableOne
import pandas as pd
import matplotlib.pyplot as plt
import tabulate
url="https://raw.githubusercontent.com/tompollard/tableone/master/data/pn2012_demo.csv"
data=pd.read_csv(url)
overall_table = TableOne(data, label_suffix=True)
x = overall_table.tableone.reset_index()
t = tabulate.tabulate(x, tablefmt="grid", headers=['isnull', 'overall'],showindex=False)
print(t)
+-------------------+------+----------+--------------+
| | | isnull | overall |
+===================+======+==========+==============+
| n | | | 1000 |
+-------------------+------+----------+--------------+
| Age, mean (SD) | | 0 | 65.0 (17.2) |
+-------------------+------+----------+--------------+
| SysABP, mean (SD) | | 291 | 114.3 (40.2) |
+-------------------+------+----------+--------------+
| Height, mean (SD) | | 475 | 170.1 (22.1) |
+-------------------+------+----------+--------------+
| Weight, mean (SD) | | 302 | 82.9 (23.8) |
+-------------------+------+----------+--------------+
| ICU, n (%) | CCU | 0 | 162 (16.2) |
+-------------------+------+----------+--------------+
| ICU, n (%) | CSRU | | 202 (20.2) |
+-------------------+------+----------+--------------+
| ICU, n (%) | MICU | | 380 (38.0) |
+-------------------+------+----------+--------------+
| ICU, n (%) | SICU | | 256 (25.6) |
+-------------------+------+----------+--------------+
| MechVent, n (%) | 0 | 0 | 540 (54.0) |
+-------------------+------+----------+--------------+
| MechVent, n (%) | 1 | | 460 (46.0) |
+-------------------+------+----------+--------------+
| LOS, mean (SD) | | 0 | 14.2 (14.2) |
+-------------------+------+----------+--------------+
| death, n (%) | 0 | 0 | 864 (86.4) |
+-------------------+------+----------+--------------+
| death, n (%) | 1 | | 136 (13.6) |
+-------------------+------+----------+--------------+
isdupe = x.duplicated(subset='variable')
x['variable'] = x['variable'].where(~isdupe, '')
t = tabulate.tabulate(x, tablefmt="grid", headers=['isnull', 'overall'],showindex=False)
print(t)
+-------------------+------+----------+--------------+
| | | isnull | overall |
+===================+======+==========+==============+
| n | | | 1000 |
+-------------------+------+----------+--------------+
| Age, mean (SD) | | 0 | 65.0 (17.2) |
+-------------------+------+----------+--------------+
| SysABP, mean (SD) | | 291 | 114.3 (40.2) |
+-------------------+------+----------+--------------+
| Height, mean (SD) | | 475 | 170.1 (22.1) |
+-------------------+------+----------+--------------+
| Weight, mean (SD) | | 302 | 82.9 (23.8) |
+-------------------+------+----------+--------------+
| ICU, n (%) | CCU | 0 | 162 (16.2) |
+-------------------+------+----------+--------------+
| | CSRU | | 202 (20.2) |
+-------------------+------+----------+--------------+
| | MICU | | 380 (38.0) |
+-------------------+------+----------+--------------+
| | SICU | | 256 (25.6) |
+-------------------+------+----------+--------------+
| MechVent, n (%) | 0 | 0 | 540 (54.0) |
+-------------------+------+----------+--------------+
| | 1 | | 460 (46.0) |
+-------------------+------+----------+--------------+
| LOS, mean (SD) | | 0 | 14.2 (14.2) |
+-------------------+------+----------+--------------+
| death, n (%) | 0 | 0 | 864 (86.4) |
+-------------------+------+----------+--------------+
| | 1 | | 136 (13.6) |
+-------------------+------+----------+--------------+
The tabulate method was added in 0.6.4:
# import libraries
from tableone import TableOne
import pandas as pd
# load sample data into a pandas dataframe
url="https://raw.githubusercontent.com/tompollard/tableone/master/data/pn2012_demo.csv"
data=pd.read_csv(url)
table = TableOne(data, label_suffix=True)
print(overall_table.tabulate(tablefmt = "fancygrid"))
outputs:
╒═══════════════════╤══════╤═══════════╤══════════════╕
│ │ │ Missing │ Overall │
╞═══════════════════╪══════╪═══════════╪══════════════╡
│ n │ │ │ 1000 │
├───────────────────┼──────┼───────────┼──────────────┤
│ Age, mean (SD) │ │ 0 │ 65.0 (17.2) │
├───────────────────┼──────┼───────────┼──────────────┤
│ SysABP, mean (SD) │ │ 291 │ 114.3 (40.2) │
├───────────────────┼──────┼───────────┼──────────────┤
│ Height, mean (SD) │ │ 475 │ 170.1 (22.1) │
├───────────────────┼──────┼───────────┼──────────────┤
│ Weight, mean (SD) │ │ 302 │ 82.9 (23.8) │
├───────────────────┼──────┼───────────┼──────────────┤
│ ICU, n (%) │ CCU │ 0 │ 162 (16.2) │
├───────────────────┼──────┼───────────┼──────────────┤
│ │ CSRU │ │ 202 (20.2) │
├───────────────────┼──────┼───────────┼──────────────┤
│ │ MICU │ │ 380 (38.0) │
├───────────────────┼──────┼───────────┼──────────────┤
│ │ SICU │ │ 256 (25.6) │
├───────────────────┼──────┼───────────┼──────────────┤
│ MechVent, n (%) │ 0 │ 0 │ 540 (54.0) │
├───────────────────┼──────┼───────────┼──────────────┤
│ │ 1 │ │ 460 (46.0) │
├───────────────────┼──────┼───────────┼──────────────┤
│ LOS, mean (SD) │ │ 0 │ 14.2 (14.2) │
├───────────────────┼──────┼───────────┼──────────────┤
│ death, n (%) │ 0 │ 0 │ 864 (86.4) │
├───────────────────┼──────┼───────────┼──────────────┤
│ │ 1 │ │ 136 (13.6) │
╘═══════════════════╧══════╧═══════════╧══════════════╛
It would be good to left align the index columns in the dataframe. See discussion on how to achieve this with styler at: https://github.com/pandas-dev/pandas/issues/39602
Currently the columns are centered when rendered in a notebook, which looks awkward:
