Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/oribarilan/workshop_genai


https://github.com/oribarilan/workshop_genai

Last synced: 4 days ago
JSON representation

Awesome Lists containing this project

README 

        
# Applied GenAI Workshop



This is a workshop on Applied Generative AI. We will focus on practical application of LLMs using LangChain.

The content of this repository is not meant to be self-served. It is meant to be used in conjunction with the workshop.

Enjoy the workshop!



## Azure Resources



We will use Azure OpenAI for this workshop.

Make sure you have a valid Azure subscription with credits, and an Azure OpenAI resource set up.

Once you have the set, you can follow the instructions below to deploy a model.



1. Go to [Azure OpenAI Studio](https://oai.azure.com/portal)

2. Click "Deployments"

3. Click "Create new deployment"

4. select "gpt-35-turbo" as the model (don't worry, you can deploy other models later)



## Installation



Throughout the workshop, we will use dev-containers to minimize the setup time and ensure alignment between participants.

You may choose to avoid dev-containers and set up your local environment. 

However, I will have limited capacity to help you with the setup during the workshop.

If you do choose to set up your local environment, please follow the manual instructions in the next section below.



### Dev Containers (Recommended)

1. Docker

    

    a. Install Docker Desktop: https://www.docker.com/products/docker-desktop

    

    b. Validate the installation by running the following command in your terminal:

    ```bash

    docker --version

    ```

    c. (Optional) Get to know docker a bit using this walkthrough https://docs.docker.com/guides/walkthroughs/run-a-container/



2. Visual Studio Code



    a. Install Visual Studio Code: https://code.visualstudio.com/



    b. Windows Only step

    

    Add this to your user-settings json file: `"dev.containers.forwardWSLServices": false`

    

    Or, using the GUI: `ctrl+shift+p` --> `Preferences: Open User Settings` --> enter this in the search box: `"dev.containers.forwardWSLServices"` and uncheck it. restart VS Code afterwards.



    Consider turning this back to `true` after the workshop, if required.



    c. Install the Dev-Containers extension: https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-containers



    d. Clone this repository to your local machine.



    e. In Visual Studio Code, open the command pallette (Cmd/Ctrl+Shift+P) and run the command dev "Dev Containers: Open Folder in Container"



    f. In the small pop-up, click the `(show logs)` link to enjoy the show. 😄



    g. Wait until the container is built and the workspace is opened in the container.



3. That's it! (The magic of dev-containers 😉)



### Manual Installation (Not recommended)



Skip this if you are using dev-containers.



1. Python

    

    a. You will need a [Python 3.9](https://www.python.org/downloads/) environment. You can install it globally or use a virtual environment, but make sure you are using Python 3.9.

    If you do this manually, I strongly suggest using [pyenv](https://github.com/pyenv/pyenv)

    

    b. Activate your python 3.9 environment and validate:

    ```bash

    python --version

    ```



2. Jupyter Notebook



    a. Install [Jupyter Notebook](https://jupyter.org/install)



    b. Validate:

    ```bash

    jupyter --version

    ```



3. Install the required Python packages



    a. Execute

    ```bash

    pip install -r requirements.txt

    ```



    b. Validate:

    ```bash

    python -c "import langchain; print(langchain.__version__)"

    ```



4. Install the required VS Code extensions (found in `devcontainer.json`)



5. Just command runner



    a. Install [Just](https://github.com/casey/just)



    b. Validate:

    ```bash

    just --version

    ```



### Setup

1. Clone the repository

2. Execute `cp .env.template .env`

3. Fill in the `.env` file with your OpenAI API key and endpoint, taken from [Azure OpenAI Studio](https://oai.azure.com/portal)



These appear in the "Keys and Endpoint" tab, under "Resource Management" in the Azure portal.