Data Visualization, Data Science and AI by Alan Jones

Articles

• AI for BI: Building a Business Information Report with CrewAI and OpenAI

  

  How to create a simple BI report directly from data

• Hands on with OpenAI Swarm

  

  Build multi-agent systems with Swarm, a back-to-basics framework

• Hands on Data Visualization with Google Mesop

  

  Google Mesop is an easy-to-use Python UI framework. We see how to use it to create Data Visualization app with Plotly.

• How to Build a Private AI Agent with Ollama and LlamaIndex

  

  Running a local LLM is a no-cost alternative to using commercial APIs

• AI Agents vs. AI Pipelines - A Practical Guide to Coding Your LLM Application

  

  We use CrewAI to create apps that demonstrate how to choose the right architecture for your LLM application

• How to Construct a Multipage Data Science Web App in Python with Taipy

  

  Taipy supports easy navigation between pages in Python web apps - we create a simple CO2 emissions app

• From Data to Visualization with the OpenAI Assistants API and GPT-4o

  

  We explore the Code Completion tool from OpenAI’s Assistants API to create visualizations directly from data

• How to Build a ReAct AI Agent with Claude 3.5 and Python

  

  We present a Reason+Act agent that iteratively reasons and gathers information from external tools before providing an answer.

• Streamlit Supports 5 Important Data Visualization Libraries - Which to Choose?

  

  We code examples using Altair, Bokeh, Plotly, Pandas Plot and Matplotlib, to illustrate the pros and cons of each one

• stlite -  How to Run Your Streamlit Apps in the Browser

  

  stlite is a browser-based implementation of Streamlit that can run apps on a web page without having to deploy to a Streamlit server

• ggplot - Grammar of Graphics in Python with Plotnine

  

  Do you wish that Python could emulate the superb visualizations that ggplot gives you in the R language? Well it can. Plotnine is a powerful graphics library for great visualizations and based on ggplot

• Why I'm Using VSCode for Jupyter Notebooks

  

  VSCode is a great Python editor and, as I accidentally discovered, good for Jupyter Notebooks, too

• Visualizing Health Risk

  

  The press can very easily get it wrong when dealing with health risk data. Data visualizations can help.

• Using Streamlit's Chat Elements, the Doctor is in

  

  We use Streamlit's chat elements to reproduce the famous 1960s Rogerian-psychologist-bot ELIZA

• Topical Plots - Global Warming Heatmaps

  

  Which chart will get the message across best

• Streamlit and MongoDB - Storing Your Data in the Cloud

  

  Deploying your Streamlit app to the Cloud means that any data that you create with that app disappears when the app terminates - unless you use third-party storage like the NoSQL database MongoDB

• Simple Surveys with Streamlit

  

  Streamlit's user interface components made constructing simple surveys easy. We create a simple survey with Streamlit, present it and show the results.

• SQL, Pandas or Both - Analysing the UK Electoral System

  

  Pandas is great for analysing and plotting data but should you store your data in a database and select it with SQL. Let's take a look at some common operations using Pandas and SQL and see how they compare

• Pyscript Evolves - How The New Version Stacks Up

  

  We build a data viz app with the new faster PyScript

• PyScript v. Flask - How to Create a Python App in the Browser or on a Server

  

  PyScript lets you create web apps in Python without the need for a server. Flask is a Python web app framework for making server-based apps. We write the same simple app using both

• Organize Your Data Science Projects with PPDA - a Case Study

  

  Define your problem, develop a plan, find the data, analyze the data and then communicate your conclusions

• Mean, Median and Mode - What Are They and When Should You Use Them?

  

  You probably remember Mean, Median, and Mode from high school stats classes but they are often misused. We look at how you should use them with Python and Pandas examples.

• Mastering the Powerful New Assistants API for Data Analyst

  

  OpenAI's Assistants API lets us create AI assistants which leverage tools that can operate on user-provided data

• How to use Ploty with PyScript

  

  PyScript doesn't support Plotly directly but there is a simple way around it

• How to Deploy Plotly Graphics to a Simple Static Web Page

  

  Plotly has a native method to create basic web pages but with a little more effort, web templates get you a much better result

• How to Create a Simple GIS Map with Plotly and Streamlit

  

  Plotly map functions combined with Streamlit user interface components provide a way of creating GIS-style dashboards

• How to Create a Grid Layout in Streamlit

  

  We present a method to programmatically create a grid layout in Streamlit with a World population data demonstration app

• How to Build a Gallery of Streamlit Apps as a Single Web App

  

  Download a free template, pop in multiple Streamlit apps and you've got a multiple Streamlit apps on a single web page

• How to Build Waterfall Charts with Plotly Graph Objects

  

  Plotly Express doesn't implement waterfall charts but we can create a helper function that utilises Plotly Graph Objects instead

• How Not to Lie with Charts

  

  You can use data visualization to inform, or misinform. We look at the things to avoid if you want to do the former and not the latter.

• How I used ChatGPT to Build a Streamlit Dashboard App

  

  Using CO² emission data I created a dashboard app in Streamlit with ChatGPT and no coding

• Flapjax - Data Visualization on the Web with Plotly and Flask

  

  Build a data visualisation web page with Plotly and Flask, and make it interactive with some UI components

• Data Visualization on the Web with Plotly and PyScript

  

  Put your data visualisations on the web with HTML templates — no server-side code required

• Data Analysis with ChatGPT and Jupyter Notebooks

  

  The conversational way of generating code with ChatGPT works well with the cell structure of Jupyter Notebooks. There is a natural workflow when using ChatGPT with Jupyter Notebooks which makes generating a code for a Jupyter Notebook using ChatGPT surprisingly easy and satisfying.

• Create an Interactive Web App with PyScript and Pandas

  

  PyScript allows us to create a serverless web application with HTML and Python as the scripting language.

• Create an AI-Generated Streamlit Application with an OpenAI Assistant

  

  We use the OpenAI Playground and code completion tools to create a Streamlit app with interactive Plotly charts

• ChatGPT Prompting for Coders - a Prototyping Approach

  

  A suitable prompting technique can make the development of usable code with ChatGPT quite painless. Here we prototype a Streamlit app for tracking cryptocurrency.

• An Interactive Web Dashboard with Plotly and Flask

  

  To create a truly interactive app with Dash you need to use callbacks. You can achieve the same thing with Plotly and Flask.

• Altair - Using the Powerful Vega-Lite 'Grammar of Graphics'

  

  Using the Altair library for Python we can develop compelling data visualizations based on a grammar of graphics and implement them in Streamlit

• A Multi-page Interactive Dashboard with Streamlit and Plotly

  

  https://towardsdatascience.com/a-multi-page-interactive-dashboard-with-streamlit-and-plotly-c3182443871a

• 4 Ways to Create a Multi-Page Streamlit App

  

  Streamlit may not have been designed for full blown web sites but it is fairly straightforward to create multiple pages in a single app

• 2023 was the Hottest Summer Ever

  

  We look at how to visualize the hottest-ever recorded temperatures over the period of June through August 2023

• 12 Essential Visualizations and How to Implement Them - part 2

  

  We look at how to create the 12 most useful graphs and charts with Python, Matplotlib and Streamlit.

• 12 Essential Visualizations and How to Implement Them - part 1

  

  We look at how to create the 12 most useful graphs and charts with Python, Matplotlib and Streamlit.