Local Installation: MacOS

1. Prepare Python Environment

You will first install Miniconda, then create a Python 3.11.3 virtual environment.

Step 1: Install Miniconda

1. Download the correct installer for your Mac:

   • Apple Silicon (M1/M2/M3, ARM64): https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh

   • Intel (x86_64): https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-x86_64.sh

2. Open Terminal:

   • Press Command (⌘) + Space, type Terminal, and press Enter

3. Run the installer for Miniconda:

   • Apple Silicon (ARM64):

     bash Miniconda3-latest-MacOSX-arm64.sh
     

   • Intel (x86_64):

     bash Miniconda3-latest-MacOSX-x86_64.sh
     

     • Follow the prompts and agree to the license terms.

     • Allow the installer to initialize Miniconda (usually by adding it to your shell’s PATH).

4. Restart Terminal and verify installation.

   • Open Terminal:

     • Press Command (⌘) + Space, type Terminal, and press Enter

   • In the terminal window, type

     conda --version
     

   • You should see the conda version displayed. If not, check your PATH settings.

Step 2: Create and Activate a Python 3.11.3 Virtual Environment

1. Open Terminal (see Step 1.2).

2. Create a new Python 3.11.3 conda environment by typing the following into terminal:

   conda create -n aihqenv python=3.11.3
   

   note that aihqenv is the name of your environment. You can replace it with any name of your choosing.

3. Activate the environment:

   conda activate aihqenv
   

   Your prompt should now show (aihqenv) to indicate the environment is active:

2. Install automated-AIHQ

Step 1: Download the automated-AIHQ

1. Visit the repository page: https://github.com/lyulouisa/Website_AIHQ_rating

   • At the top-right area of the page, just above the list of files, you’ll see a green button labeled

     <>Code

   • Click the <> Code button, and in the menu that appears, choose “Download ZIP.” This will download a ZIP file of the website folder onto your computer with the name “Website_AIHQ_rating-main.zip”

2. Extract the ZIP file:

   • Once the ZIP file finishes downloading, open your Downloads folder (or wherever your browser saves files).

   • Look for a file named Website_AIHQ_rating-main.zip.

   • Double-click the ZIP file. Your Mac will automatically unzip it into a folder with the same name (but without .zip at the end). Alternatively, you can right-click the zipped folder and select Open With → Archive Utility to unzip the folder.

   • You should now see a folder called Website_AIHQ_rating-main. This is the automated-AIHQ tool.

3. Download the fine-tuned Flan-T5-large model (you can ignore this step if you do not intend to use this model)

   • There are three options to download the model:

     1. Zenodo: https://zenodo.org/records/16730672 (download the zipped folder)

     2. Dropbox: https://www.dropbox.com/scl/fi/8knvlq83r9j031axqiqq7/AIHQ_rating.zip?dl=0

     3. Hugging Face: https://huggingface.co/lyulouisaa/flant5-finetuned-aihqrating You will need to download each file directly and save it in a folder titled: flant5-large-finetuned

   • Once the download is complete, double-click to unzip. You should see a folder named flant5-large-finetuned

4. Place the folder in the Website_AIHQ_rating folder:

   • Drag the flant5-large-finetuned folder into the Website_AIHQ_rating-main folder.

   • When you are done, your folder structure should look like:

5. Open Terminal:

   • Press Command (⌘) + Space, type Terminal, and press Enter to open the Terminal

   • Navigate to the automated-AIHQ folder:

     • In the command line window, type cd /path/to/Website_AIHQ_rating-main and press Enter

     • Replace /path/to/Website_AIHQ_rating-main with the location of the folder where you unzipped the files.

6. Run the script:

   • In the same command line window, type python main.py and press Enter

7. Access the automated-AIHQ interface:

   • Once the installation is complete, the command line will display a link, e.g.,

     http://127.0.0.1:5005

   • Open this link in your web browser. The automated-AIHQ interface will appear.

The above image shows the main interface of automated-AIHQ. Here, you begin by selecting a model and uploading your CSV file for rating.

Launching Automated-AIHQ after successful installation

If you have already installed the tool, you just need to follow the following steps to open the tool.

1. Open Terminal:

   • Press Command (⌘) + Space, type Terminal, and press Enter to open the Terminal

   • Navigate to the automated-AIHQ folder:

     • In the command line window, type cd /path/to/Website_AIHQ_rating-main and press Enter

     • Replace /path/to/Website_AIHQ_rating-main with the location of the folder where you unzipped the files.

2. Run the script:

   • In the same command line window, type python main.py and press Enter

3. Access the automated-AIHQ interface:

   • Once the installation is complete, the command line will display a link, e.g.,

     http://127.0.0.1:5005

   • Open this link in your web browser. The automated-AIHQ interface will appear.

Troubleshooting

1. Version mismatch (NumPy/Pandas) error when running python main.py: This is typically due to incompatible package versions. The included installation.py pins:

   • pandas==1.5.3

   • numpy==1.24.3

   If you still encounter errors, ensure you’re inside the aihqenv environment and then run:

   python -V
   

   Confirm it shows Python 3.11.3. Next, reinstall the pinned packages:

   python -m pip uninstall -y pandas
   python -m pip install pandas==1.5.3
   python -m pip install numpy==1.24.3
   

   Then try

   python main.py
   

2. Model folder not found:

   Double-check that flant5-large-finetuned is inside your website folde e.g.:

   Website_AIHQ_rating-main/flant5-large-finetuned
   

Video

Click here for a video showing you how to install automated-AIHQ on your Mac step by step.