Skip to content

wjbmattingly/streamlit-pandas

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Streamlit pandas logo

Streamlit Pandas

Streamlit Pandas is a component for the Streamlit library. It allows users to load a Pandas DataFrame and automatically generate Streamlit widgets in the sidebar. These widgets trigger filtering events within the Pandas DataFrame.

Support

Current support only exists for DataFrame columns with strings and numbers (int64 and float64). A future update will include support for time-series data.

By default, string data generates a text_input Streamlit widget, while numerical data creates sliders with ranges preset to the minimum and maximum values for that column. Users can pass a custom dictionary for handling specific types of data, where each key is the column in the DataFrame and the value is the streamlit widget type.

Sample of a custom dictionary:

create_data = {"Name": "text",
                "Sex": "multiselect",
                "Embarked": "multiselect",
                "Ticket": "text",
                "Pclass": "multiselect"}

The current version only supports: text, multiselect, and select.

Installation

  1. First, install Streamlit
pip install streamlit
  1. Next, install Pandas
pip install pandas
  1. Install Streamlit Pandas
pip install streamlit-pandas

Usage

import streamlit as st
import pandas as pd
import streamlit_pandas as sp

@st.cache_data
def load_data():
    df = pd.read_csv(file)
    return df

file = "../data/titanic.csv"
df = load_data()
create_data = {"Name": "text",
                "Sex": "multiselect",
                "Embarked": "multiselect",
                "Ticket": "text",
                "Pclass": "multiselect"}

all_widgets = sp.create_widgets(df, create_data, ignore_columns=["PassengerId"])
res = sp.filter_df(df, all_widgets)
st.title("Streamlit AutoPandas")
st.header("Original DataFrame")
st.write(df)

st.header("Result DataFrame")
st.write(res)

This will generate the following application:

Streamlit-Pandas demo application