{"id":26585422,"url":"https://github.com/zseta/scylladb-feast","last_synced_at":"2026-05-04T19:32:56.196Z","repository":{"id":283189649,"uuid":"950954377","full_name":"zseta/scylladb-feast","owner":"zseta","description":null,"archived":false,"fork":false,"pushed_at":"2025-03-19T01:08:03.000Z","size":23382,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-19T02:23:16.788Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/zseta.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-03-19T00:04:50.000Z","updated_at":"2025-03-19T01:08:06.000Z","dependencies_parsed_at":"2025-03-19T02:33:36.778Z","dependency_job_id":null,"html_url":"https://github.com/zseta/scylladb-feast","commit_stats":null,"previous_names":["zseta/scylladb-feast"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zseta%2Fscylladb-feast","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zseta%2Fscylladb-feast/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zseta%2Fscylladb-feast/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zseta%2Fscylladb-feast/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zseta","download_url":"https://codeload.github.com/zseta/scylladb-feast/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245090894,"owners_count":20559296,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-03-23T10:39:34.668Z","updated_at":"2026-05-04T19:32:56.190Z","avatar_url":"https://github.com/zseta.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Real-time credit scoring application with Feast \u0026 ScyllaDB\n\nThis project is based on [this](https://github.com/feast-dev/feast-aws-credit-scoring-tutorial) existing Feast sample application.\n\n![scylla feast architecture](/images/scylla-feast.jpg)\n\nThis sample project is a real-time credit scoring application example that shows you how to set up Feast with ScyllaDB Cloud as an online store and parquet files as offline store.\n\n* The primary training dataset is in `loan_table.parquet`. This file contains historic loan data with accompanying features. The dataset also contains a target variable, namely whether a user has defaulted on their loan.\n* Feast is used during training to enrich the loan table with `zipcode` and `credit history` features from other parquet files.\n* Feast is also used to serve the latest `zipcode` and `credit history` features for online credit scoring using ScyllaDB Cloud.\n\n## Requirements\n\n* Python 3\n* [ScyllaDB Cloud account](https://cloud.scylladb.com/)\n\n## Get started\n\nClone the repository:\n```\ngit clone https://github.com/zseta/scylladb-feast\ncd scylladb-feast\n```\n\nIn this tutorial, you'll do the following steps:\n1. Create a new ScyllaDB Cloud cluster\n1. Create a new keyspace for the feature store\n1. Install dependencies and configure Feast\n1. Deploy and test the feature store\n\n### Create ScyllaDB Cloud cluster\nGo to [ScyllaDB Cloud](https://cloud.scylladb.com/) and create a new cluster (either \"Free Tial\" or \"Dedicated VM\"). You can use the smallest available machine for this sample app (`t3.micro`)\n![choose machine type](images/choose-machine.png)\n\n\n### Create new keyspace\nInstall CQLSH command line tool and connect to the ScyllaDB cluster:\n```\npip install cqlsh\ncqlsh \u003cSCYLLA-CLOUD-HOST\u003e -u scylla -p \u003cPASSWORD\u003e\n```\n\nYou can get the host address, username, and password from your ScyllaDB Cloud dashboard:\n![scylla connect](/images/scylla-cloud-connect.png)\n\nCreate a new keyspace called `feast` in ScyllaDB (this keyspace will be populated by Feast):\n```\nCREATE KEYSPACE feast WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': '3'} ;\n```\n\n### Install dependencies and configure Feast\n\nCreate a new Python environment and install dependencies:\n```\nvirtualenv env \u0026\u0026 source env/bin/activate\npip install -r requirements.txt\n```\n\nNext, configure ScyllaDB Cloud as the online store for Feast. \n\n\u003e [!TIP]\n\u003e ScyllaDB is compatible with Apache Cassandra so you can use the Feast Cassandra connector with ScyllaDB\n\nOpen the feature_store.yaml file and add the host addresses, username (`scylla`), and password for ScyllaDB: \n\n```yaml\nproject: repo\n# By default, the registry is a file (but can be turned into a more scalable SQL-backed registry)\nregistry: data/registry.db\n# The provider primarily specifies default offline / online stores \u0026 storing the registry in a given cloud\nprovider: local\nonline_store:\n    type: cassandra\n    hosts:\n        - x.x.x.x\n        - x.x.x.x\n        - x.x.x.x\n    username: scylla\n    password: pass\n    keyspace: feast\nentity_key_serialization_version: 2\n```\n\nYou can get these values from the ScyllaDB Cloud dashboard.\n\n\n### Deploy and test the feature store\n\nDeploy the feature store by running `apply` from within the `feature_repo/` folder\n```\ncd feature_repo/\nfeast apply\n\n\nDeploying infrastructure for credit_history\nDeploying infrastructure for zipcode_features\n```\n\nThis commmand also generates a `registry.db` file in your `data/` folder containing Feast configuration information.\n\nNext, load features into the online store using the `materialize-incremental` command. This command loads the\nlatest feature values from a data source (parquet files, in our case) into the online store (ScyllaDB).\n\n```\nCURRENT_TIME=$(date -u +\"%Y-%m-%dT%H:%M:%S\")\nfeast materialize-incremental $CURRENT_TIME\n```\n\nThe loading process starts:\n```\n 35%|███████████████████▋                                     | 9965/28844 [01:20\u003c02:39, 118.37it/s]\n```\n\nWhen it's completed, train the model using `run.py` (in the root folder of the repository)\n```\ncd ..\npython run.py\n```\n\nThe script returns the result of a single loan application:\n```\nLoan rejected!\n```\n\nYou can also run `feast ui` to explore your features:\n```\nfeast ui\n```\n\nOpen http://0.0.0.0:8888\n![feast ui](/images/feast-ui.png)\n\n### Interactive demo\nRun the Streamlit app locally:\n```\nstreamlit run app.py\n```\n\nGo to http://localhost:8501\n\n![streamlit app](/images/streamlit_app.png)\n\nYou can modify the variables on the left sidebar then the app will automatically run the prediction using the model.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzseta%2Fscylladb-feast","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzseta%2Fscylladb-feast","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzseta%2Fscylladb-feast/lists"}