We recommend checking out the following before getting started on PA3:

• The Week 3 videos and slides.
• The Week 3 group work and slides on Information Retrieval.
• Chris Manning's Canvas Videos on Information Retrieval.
• The Manning, Raghavan, Schütze readings on Boolean Retrieval, Term vocabulary and postings lists, Scoring, term weighting, and the vector space model, and Evaluation in Information Retrieval

You can get started using the same steps you followed for the previous PAs. To recap, you'll want to:

1. Open a terminal (terminal for macOS and Linux, Ubuntu for Windows for Windows, or SSH into Rice/Myth) the same way you did for the previous PAs.

2. Clone the git repository for PA3 (this repository) into a folder of your choice by typing this in your terminal:

    git clone https://github.com/cs124/pa3-ir.git
   

3. Enter the project root directory and activate your cs124 conda environment by doing:

    cd pa3-ir
    conda activate cs124
   

   You should now see (cs124) in front of your shell prompt.

   You'll need to do this every time you open a new terminal and re-start your notebook server. You should have already created this conda environment as part of PA0. If not, and you encounter an error, please go back and follow the instructions there.

4. Start up your jupyter notebook server

    jupyter notebook
   

5. A window should open automatically in your default browser. If it didn't, the terminal output should contain a URL you can use to open the notebook in a browser of your choice.

6. From the Jupyter notebook file explorer window that opens, click on the pa3.ipynb file to open and edit it.

1. Go to colab.research.google.com. If prompted to open a notebook, hit cancel for now. You should double-check that you are logged in to your Stanford Google account. If not, you can switch accounts in the top right.

2. Now go to File->Open Notebook. Go to the GitHub tab. It will ask you to log in to your GitHub account (if you don't have one it is easy to make one). Once you've done that, copy and paste the URL: https://github.com/cs124/pa3-ir into the search box and hit enter. It should show:

        Repository: cs124/pa3-ir
        Branch:  Main
   

   Click on pa3.ipynb below to load the notebook.

Rice is the name of a group of machines made available for student use by Stanford. Myth machines are similar, but specifically maintained by and for the computer science department. For our purposes, they are interchangeable, so you should feel free to use whichever one you prefer.

0. [WINDOWS ONLY] If you are using Windows, you have two options. If you were able to install Ubuntu for Windows according to the instructions above, you can use that for your terminal/shell to SSH into Rice/Myth (although, if you were able to get Ubuntu for Windows set up we would strongly recommend that you try to get things working locally instead of using Rice/Myth). If you weren't able to install Ubuntu for Windows, you can instead download and install PuTTY here to use as an SSH client. Install and run it.

1. SSH into Rice/Myth.

   • [macOS/Linux/Ubuntu for Windows]

        ssh [Your SUNet]@rice.stanford.edu
        # OR
        ssh [Your SUNet]@myth.stanford.edu
     

   • [Windows] Run PuTTY and when prompted, enter rice.stanford.edu or myth.stanford.edu in the box labeled Host Name. Do not change any of the other settings. Click open to open a connection, which will cause a terminal window to pop up. You will then be prompted to log in.

   You'll be prompted for your Stanford SUNet account password. You will also likely be asked to authenticate using 2-factor authentication.

2. Clone the git repository for PA3 in a location of your choice:

    git clone https://github.com/cs124/pa3-ir.git
   

3. Enter the project root directory and activate your cs124 conda environment (you should have already created this as part of PA0. If not, please go back and follow the instructions there):

    cd pa3-ir
    conda activate cs124
   

4. Start up your jupyter notebook server without a browser.

    jupyter notebook --no-browser --port=[pick a number from 1025 to 65535]
   

5. You now need to set up port forwarding from Rice/Myth to your local machine:

   • [macOS/Linux/Ubuntu for Windows] Open a new terminal window on your local machine (NOTE: don't close the one on Rice/Myth, that one needs to stay open) and enter the following command:

        ssh -NL [local port you'd like to use]:localhost:[port number you picked on rice/myth] [SUNet]@[rice/myth].stanford.edu
     

   If it worked correctly, nothing should happen and your terminal will appear to hang. This is fine, you should keep it open to maintain the tunnel.

   • [Windows] Go back to your PuTTY window. In the left-side navigation window and select Connection->SSH->Tunnels. In the right-hand side window, there will be a section labeled "Add new forwarded port".

   Under Source port, type the local port you'd like to forward to. Destination, type:

    [rice/myth].stanford.edu:[port your Jupyter notebook server is running on]
   

   Then, click "Add".

6. Finally, try opening a browser of your choice (i.e. Chrome, Firefox, Safari) and opening the URL:

    localhost:[local port number you chose]
   

   If everything worked correctly, you'll see the Jupyter file explorer. It may prompt you for an authentication token. If it does, go back to the terminal window that is running the Jupyter notebook server (on Rice/Myth). In the output, you should see some URLs with a portion that looks like "token=...". Copy the token from there and paste it into your browser where requested.

7. From the Jupyter notebook file explorer window that opens, click on the pa3.ipynb file to open it. All your implementation will be done directly in the notebook.

1. You can run the cell at the bottom of the Jupyter notebook to zip up your solution for you. It should generate a zip file submission.zip.

2. Upload the zip file as your solution to the PA3 IR assignment in Gradescope (http://www.gradescope.com).

1. You can run the cell at the bottom of the Jupyter notebook to zip up your solution for you. It should generate a zip file submission.zip.

2. To access the file easily from Windows (which you'll need to do if you want to upload it to Gradescope), you'll probably want to move or copy it to an easily accessible location in your Windows filesystem. The C drive in Windows is accessible in Ubuntu at /mnt/c, so you should copy it to some path there, like /mnt/c/Users/[your Windows username]/Desktop, which is the path to your Windows desktop directory. You can do this from your Ubuntu terminal by doing:

    cp /path/to/submission.zip /mnt/c/Users/[your Windows username]/Desktop
   

3. Upload the zip file as your solution to the PA3 IR assignment in Gradescope (http://www.gradescope.com).

1. Once you've saved all your changes, go to File->Download .ipynb to download your notebook file to your local machine.

2. If your solution required any extra files, make sure they were located in pa3-ir/deps. Go to the file explorer on the left-hand side and download those as well.

3. Once you have pa3.ipynb and any deps files you need saved locally, create a zip file (either from your OS's file explorer or the terminal), making sure that it has the following structure:

    deps/
    pa3.ipynb
   

4. Upload the zip file as your solution to the PA3 IR assignment in Gradescope (http://www.gradescope.com).

1. You can run the cell at the bottom of the Jupyter notebook to zip up your solution for you. It should generate a zip file submission.zip.

2. You will then need to download/copy the zip file from Rice/Myth to your local machine.

   • [macOS/Linux/Ubuntu for Windows]

     Open a terminal on your local machine and run the following command to copy the submission zip file from Rice/Myth to your local machine.

        scp [SUNet ID]@[myth/rice].stanford.edu:/path/to/submission.zip /local/path/to/copy/to
     

     If you're using Ubuntu for Windows, we recommend choosing as your local path something starting with /mnt/c, as this is the UNIX path that corresponds to your Windows filesystem. This allows you to download the zip file to a location where it can be easily accessed directly from windows. For example, if you chose /mnt/c/Users/[your Windows username]/Desktop as your destination, the file will be saved to your Windows Desktop folder.

   • [Windows] PSCP is another utility that should have been installed automatically when you installed PuTTY earlier. Find and run it. In the terminal window, you can run the command:

        pscp [Your SUNet]@rice.stanford.edu:/path/to/submission.zip c:\temp\submission.zip
     

   to download the file from Rice/Myth to your local machine. You should now be able to access the file locally at c:\temp\submission.zip

3. Upload the zip file as your solution to the PA3 IR assignment in Gradescope (http://www.gradescope.com).