Skip to content

Latest commit

 

History

History
104 lines (88 loc) · 2.75 KB

Installation.md

File metadata and controls

104 lines (88 loc) · 2.75 KB

Installation for Ubuntu

Setting up the environment

To install the Protein Graph Database on Ubuntu, follow these steps:

  1. Clone the repository. To access the repo, you need to add your SSH key to your GitHub account. Refer to this guide for more information.

    git clone [email protected]:BackofenLab/protein-graph-database.git
cd protein-graph-database/

  2. Install the required software, including Anaconda, Java, and Neo4j. Refer to Requirements.mk for details.

    make requirements

  3. Create the conda environment.

    make env
conda activate pgdb

Dummy Data Setup

To automate the download of dummy data from Google Drive and the insertion into the database, follow these steps:

  1. Run the following command to set up the dummy data:

    make dummydata

Executing the above command will seamlessly handle the installation of the gdown tool, download the dummy data, and insert it into the database. Gdown is a helper tool to automate the download from Google Drive.

Create a Neo4J Account

To create a Neo4j account, follow these steps:

  1. Start Neo4j with the following command: 
    make neo4j
  2. Open localhost:7474 in your web browser.
    • Login with the following credentials:
      • Username: neo4j
      • Password: neo4j
    • Change the password to: pgdb.

Set Credentials

  1. Create the file protein-graph-database/config.yml
  2. Insert the following content: 
    # Neo4j Database Credentials
neo4j:
  host: "localhost"
  port: 7687
  username: "neo4j"
  password: "pgdb"

Installing and using ollama

  1. Download and install ollama: 
    curl -fsSL https://ollama.com/install.sh | sh
  2. Make sure ollama for python is installed: 
    pip install ollama
  3. Start ollama as a system: 
    sudo systemctl start ollama
  4. To check if ollama was correctly started and the gpu recognized check logs: 
    journalctl -e -u ollama
  5. Add environmental variables (such as the one to keep the model alive and loaded in memory): 
    sudo systemctl edit ollama.service

    [Service]
Environment="OLLAMA_KEEP_ALIVE=-1"
  6. Restart ollama and reload systemd: 
    systemctl daemon-reload
systemctl restart ollama
  7. Soft start llama3 to keep in memory: 
    ollama run llama3 ""
  8. Check if the model was correctly loaded and is using the gpu: 
    ollama ps