Skip to main content

4.7 Dashboards - Streamlit

Streamlit is an open-source Python library that allows data scientists and developers to create interactive, web-based applications with ease. Built with a focus on machine learning and data analysis, Streamlit offers an intuitive way to turn data scripts into shareable web applications without requiring extensive web development skills.

Why Streamlit?​

  • Simplicity: You can create a web application with just a few lines of Python code. No need to juggle between different files or learn HTML, CSS, or JavaScript.
  • Interactivity: Widgets like sliders, buttons, and input boxes allow users to interact with the data and visualizations.
  • Data Integration: Streamlit supports a variety of data sources and integrates seamlessly with popular data science libraries like Pandas, NumPy, and Matplotlib.
  • Live Reloading: As you edit your Streamlit script, the application automatically updates, offering a quick feedback loop during development.
  • Deployment and Sharing: With platforms like Streamlit sharing, deploying your app to the world becomes a cakewalk.

Building a Basic Streamlit App​

Installation​

  1. Install Streamlit:

    pip install streamlit

  2. Run your app: 

    streamlit run your_app_script.py

Examples​

A basic Streamlit application might look something like this:

import streamlit as st
import pandas as pd
import numpy as np

# Title of the App
st.title('My First Streamlit App')

# Add a widget (e.g., a slider)
number = st.slider("Select a number", 1, 100, 50)

st.write(f"You selected the number {number}")

# Load and display some data
data = pd.DataFrame({
  'numbers': np.random.randn(100),
})

st.line_chart(data)

To run a Streamlit app, you would save your script (e.g., app.py) and then run streamlit run app.py in your terminal.

Here is a more advanced example, that has some capabilities highlighted such as 'data caching' to enable for quicker reloads:


import streamlit as st
import pandas as pd
import numpy as np

st.title('Uber pickups in NYC')

DATE_COLUMN = 'date/time'
DATA_URL = ('https://s3-us-west-2.amazonaws.com/'
            'streamlit-demo-data/uber-raw-data-sep14.csv.gz')

@st.cache_data
def load_data(nrows):
    data = pd.read_csv(DATA_URL, nrows=nrows)
    lowercase = lambda x: str(x).lower()
    data.rename(lowercase, axis='columns', inplace=True)
    data[DATE_COLUMN] = pd.to_datetime(data[DATE_COLUMN])
    return data

data_load_state = st.text('Loading data...')
data = load_data(10000)
data_load_state.text("Done! (using st.cache_data)")

if st.checkbox('Show raw data'):
    st.subheader('Raw data')
    st.write(data)

st.subheader('Number of pickups by hour')
hist_values = np.histogram(data[DATE_COLUMN].dt.hour, bins=24, range=(0,24))[0]
st.bar_chart(hist_values)

# Some number in the range 0-23
hour_to_filter = st.slider('hour', 0, 23, 17)
filtered_data = data[data[DATE_COLUMN].dt.hour == hour_to_filter]

st.subheader('Map of all pickups at %s:00' % hour_to_filter)
st.map(filtered_data)

Integrating with Other Libraries​

One of the strengths of Streamlit is its integration with other libraries. For instance, if you've created a visualization with Matplotlib, Seaborn, or Plotly, you can easily display it in a Streamlit app. Similarly, machine learning models trained with libraries like Scikit-learn or TensorFlow can be integrated and showcased within your Streamlit application.

Limitations and Considerations​

While Streamlit offers a fast and intuitive way to build data applications, it's essential to note a few limitations:

  • State Management: Managing state (like remembering user inputs across interactions) can be tricky. There are workarounds, but it's not as straightforward as some other frameworks.
  • Custom Styling: While Streamlit provides a clean and modern look out of the box, extensive custom styling requires more effort.

Streamlit bridges the gap between data analysis and web application, providing an efficient and elegant way to share insights, models, and visualizations with a broader audience. As with any tool, it's essential to understand its strengths and limitations to determine if it's the right fit for your project.