I have a large (vertically) pandas dataframe that I would like to display as a nice table with scrollbars. I can get it to display the table with all the rows but I cannot get the scrollbars to display.
def data(x):
strData = strData[['Data1','Data2','Data3']]
display(strData)
output: No vertical scrollbars
Not sure if this is what you mean, but I guess you need to set the max_rows option to None, so that pandas doesn't put a limit on the number of rows displayed:
pd.set_option("display.max_rows", None)
Update:
In [27]:
##javascript
IPython.OutputArea.auto_scroll_threshold = 10;
In[28]:
def display_():
pd.set_option("display.max_rows", None)
from IPython.core.display import display
display(df) #df must be defined up there
def display_(): pd.set_option("display.max_rows", None) from IPython.core.display import display display(df)
I'd strongly recommend dtale as a more sophisticated way to display DataFrames in a notebook. It has scrolling, but it also efficiently handles large DataFrames, and allows editing and sorting, among other things.
If you don't want an additional library (for some reason):
The other answer didn't work for me - IPython.OutputArea doesn't seem to exist any more (as far as I can tell, at least not in VSCode and based on the IPython code).
I managed to make a DataFrame scrollable with a somewhat hacky solution of generating the HTML table for the DataFrame and then putting that into a div, which can be made scrollable using CSS. In this case all we really need to do is set the height to some pixel value.
We can also set the overflow to auto and the width to fit-content so the scrollbar appears right next to the DataFrame, instead of all the way on the right of the window.
import pandas as pd
from IPython.display import display, HTML
df = pd.DataFrame([(i, i) for i in range (20)])
pd.set_option("display.max_rows", None)
# Puts the scrollbar next to the DataFrame
display(HTML("<div style='height: 200px; overflow: auto; width: fit-content'>" +
df.style.render() +
"</div>"))
# Puts the scrollbar on the right side of the window
display(HTML("<div style='height: 200px'>" + df.style.render() + "</div>"))
Demo:
height with max-height to prevent the table to take up unnecessary space when there are few items in it: display(HTML("<div style='max-height: 400px'>" + pd.DataFrame(records, columns=columns).to_html() + "</div>"))
Just pass the dataframe and observe the magic.
def table(df):
import plotly.graph_objs as go
fig = go.Figure(data=[go.Table(
header=dict(values=list(df.columns),
align='left'),
cells=dict(values=[df[i] for i in df.columns],
align='left'))
])
return fig
This is what it looks like:
Using ipydatagrid:
import string
import numpy as np
import pandas as pd
N = 5000
cols = {char:np.random.randn(N) for char in string.ascii_lowercase[:5]}
df = pd.DataFrame(cols)
from ipydatagrid import DataGrid
DataGrid(df)
