Gemini 101 Workshop

This is a workshop dedicated to introducing Google's Gemini API with Python and Streamlit. You'll learn how to generate an API key, authenticate with the API, make requests to the Gemini API, and output the response to the web with streamlit.

1. Create a Gemini API Key

We'll first start with creating a new API key.

Step 1: Open Google AI Studio and Login / Create an Account

Here's the link: https://aistudio.google.com/app/apikey

Step 2: Click on "Create API key"

This opens the modal to select which GC project to create an API in.

Step 3: Click "Create API key in new project"

If you don't have an existing project, you can create a new one. Otherwise, select the project you'd like to create the API key for.

Step 4: Copy the API key

Copy the API key and save it for later. We'll need it soon.

2. Setting up the project folder structure

Now that we've created our API key, we need to create a new folder where our Python files will be stored.

Step 1: Create a new folder

Create the folder that you want to store your Python code and files in.

Steps 2: Creating required files

Our main code folder

This is where we'll store our source code.

1. Create a new folder called: src
2. Inside the src folder, create a new file called: main.py
3. Inside the src folder, create a new folder called: utils.
4. Inside the src/utils folder, create a new file called: to_markdown.py

Done!

Creating the config file

JSON is like a dictionary in python. It's has key-value pairs. We'll store our API key in this file.

1. Create a new file called: config.json
2. Add the following content to the file:

{
  "gemini": {
    "api_key": "YOUR GEMINI API KEY"
  }
}

Done!

Coding

Now we'll actually get into the coding!

Requirements

First we need to install some libraries.

pip install -q -U google-generativeai streamlit IPython

File: to_markdown.py

We'll start with coding the to_markdown function. This function will convert a text to markdown format.

import textwrap
from IPython.display import Markdown


##
## Convert a text to markdown format
##
def to_markdown(text: str):
    text = text.replace("•", "  *")

    return Markdown(textwrap.indent(text, "> ", predicate=lambda _: True))

File: app.py

This is our main file. See src/app.py for the code.

Running the program

An example of running the program is shown below:

streamlit run src/app.py

Basically, you'll need the path to your app.py file. You can get this however you like, but if you run the python file normally, streamlit will output the proper command to run.