WTFAQ?

Starting with Own Model Creation using TrainMyAI:

1. Requirements:

  • A modern 64-bit Linux server (Tested on Ubuntu 18/20/22, Debian 10/11/12, and CentOS Stream 8/9).
  • Minimum 4 GB of memory and a 40 GB SSD drive.
  • Apache 2.x, MySQL/MariaDB 5.5 or later, and PHP 7.2 or later.
  • An OpenAI/ChatGPT API key (a paid account is recommended).

2. Installation:

  • Install necessary packages like Apache, PHP, MySQL, Poppler, catdoc, and docx2txt.
  • Download TrainMyAI using wget and extract its contents.
  • Create a data directory for TrainMyAI and ensure it’s owned by the Apache web server user.

3. Database Setup:

  • Choose a database name, user name, and password for TrainMyAI.
  • Log in to MySQL or MariaDB as the root user and create the database, user, and grant necessary privileges.

4. Configuration:

  • Copy the example configuration file and edit it.
  • Set the data directory, database credentials, and OpenAI API key in the configuration file.

5. Web Server Setup:

  • Create a symbolic link in your web server’s content directory to TrainMyAI’s html directory.
  • Ensure Apache is configured to follow symbolic links and .htaccess files.
  • Activate Apache’s rewrite module and restart the web server.

6. Cron Job:

  • Set a cron job to make regular requests to a special web page for asynchronous tasks of TrainMyAI.

7. Loader Installation:

  • Open the installation verification page in your browser.
  • Install the SourceGuardian loader as per the instructions.
  • Restart the web server and refresh the installation verification page.
  • Sign up as the first user on TrainMyAI to get full admin permissions.

Once done, you can create your first knowledge base and start adding content. Further, you can explore customization and API integration options.

 

— Was this helpful? · 👍 · 👎 —