Quick Start

This guide walks you through the quickest way to setup a recommender system for GitHub repositories based on dataset from GitRec. Make sure you have installed the following softwares at the beginning:

• Docker, nodes will run in containers.
• Docker Compose, nodes will be orchestrated by Docker Compose.

Setup Gorse

Gorse supports to be single node or multiple nodes. This section demonstrates how to setup a minimal usable Gorse single node via playground and a multiple nodes cluster via Docker Compose.

Option 1: Setup Gorse-in-one node via Playground

Use the following command to setup a Gorse all-in-one node.

curl -fsSL https://gorse.io/playground | bash

docker run -p 8088:8088 zhenghaoz/gorse-in-one --playground

Tips

For Gorse-in-one, the RESTful endpoint is http://127.0.0.1:8088.

Option 2: Setup Gorse Cluster via Docker Compose

There is an example docker-compose.yml consists of a master node, a server node and a worker node, a Redis instance, and a MySQL instance.

1. Download docker-compose.yml and the config file config.toml.

# Create a new directory
mkdir gorse
cd gorse

# Download docker-compose.yml and config.tom
wget https://raw.githubusercontent.com/zhenghaoz/gorse/release-0.4/docker-compose.yml
wget https://raw.githubusercontent.com/zhenghaoz/gorse/release-0.4/config.toml

2. Setup the Gorse cluster using Docker Compose.

docker-compose up -d

Expected outputs

Creating network "gorse_default" with the default driver
Creating gorse_worker_1 ... done
Creating gorse_master_1 ... done
Creating gorse_mysql_1  ... done
Creating gorse_server_1 ... done
Creating gorse_redis_1  ... done

3. Download the SQL file github.sql and import to the MySQL instance. This dataset consists of GitHub users, GitHub repositories and interactions between users and repositories.

# Download sample data.
wget https://cdn.gorse.io/example/github.sql

# Import sample data.
mysql -h 127.0.0.1 -u gorse -pgorse_pass gorse < github.sql

There are three kinds of interactions (called "feedbacks" in Gorse) between users and repositories in this dataset.

Feedback Type	Description
star	A user "starred" a repository in GitHub
read	A user read a repository in GitRec
like	A user "liked" a repository in GitRec

In the configuration file, "read" belongs to read_feedback_types while "star" and “like” belong to positive_feedback_types.

# The feedback types for positive events.
positive_feedback_types = ["star","like"]

# The feedback types for read events.
read_feedback_types = ["read"]

4. Restart the master node to reload imported data immediately.

docker-compose restart

Expected outputs

Restarting gorse_redis_1  ... done
Restarting gorse_mysql_1  ... done
Restarting gorse_master_1 ... done
Restarting gorse_server_1 ... done
Restarting gorse_worker_1 ... done

Get Recommendation

Gorse generate recommendations for users after feedbacks from users are inserted.

Open http://127.0.0.1:8088 in browser and the dashboard summarizes all status and data in Gorse.

Wait for all tasks except "Searching collaborative filtering model" and "Searching click-through prediction model" completed.

Suppose Bob is a frontend developers who starred several frontend repositories in GitHub. We insert his star feedback to Gorse via the RESTful API.

read -d '' JSON << EOF
[
    { \"FeedbackType\": \"star\", \"UserId\": \"bob\", \"ItemId\": \"vuejs:vue\", \"Timestamp\": \"2022-02-24\" },
    { \"FeedbackType\": \"star\", \"UserId\": \"bob\", \"ItemId\": \"d3:d3\", \"Timestamp\": \"2022-02-25\" },
    { \"FeedbackType\": \"star\", \"UserId\": \"bob\", \"ItemId\": \"dogfalo:materialize\", \"Timestamp\": \"2022-02-26\" },
    { \"FeedbackType\": \"star\", \"UserId\": \"bob\", \"ItemId\": \"mozilla:pdf.js\", \"Timestamp\": \"2022-02-27\" },
    { \"FeedbackType\": \"star\", \"UserId\": \"bob\", \"ItemId\": \"moment:moment\", \"Timestamp\": \"2022-02-28\" }
]
EOF

curl -X POST http://127.0.0.1:8087/api/feedback \
   -H 'Content-Type: application/json' \
   -d "$JSON"

Expected outputs

{
 "RowAffected": 5
}

Then, fetch 10 recommended items from Gorse. We can found frontend related repositories are recommended for Bob.

curl http://127.0.0.1:8087/api/recommend/bob?n=10

Expected outputs:

[
    "mbostock:d3",
    "nt1m:material-framework",
    "mdbootstrap:vue-bootstrap-with-material-design",
    "justice47:f2-vue",
    "10clouds:cyclejs-cookie",
    "academicpages:academicpages.github.io",
    "accenture:alexia",
    "addyosmani:tmi",
    "1wheel:d3-starterkit",
    "acdlite:redux-promise"
]