You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
*`feature_repo/feature_store.yaml` contains a demo setup configuring where data sources are
8
+
*`feature_repo/test_workflow.py` showcases how to run all key Feast commands, including defining, retrieving, and pushing features.
9
+
10
+
You can run the overall workflow with `python test_workflow.py`.
11
+
12
+
## To move from this into a more production ready workflow:
13
+
> See more details in [Running Feast in production](https://docs.feast.dev/how-to-guides/running-feast-in-production)
14
+
15
+
1. First: you should start with a different Feast template, which delegates to a more scalable offline store.
16
+
- For example, running `feast init -t gcp`
17
+
or `feast init -t aws` or `feast init -t snowflake`.
18
+
- You can see your options if you run `feast init --help`.
19
+
2.`feature_store.yaml` points to a local file as a registry. You'll want to setup a remote file (e.g. in S3/GCS) or a
20
+
SQL registry. See [registry docs](https://docs.feast.dev/getting-started/concepts/registry) for more details.
21
+
3. This example uses a file [offline store](https://docs.feast.dev/getting-started/components/offline-store)
22
+
to generate training data. It does not scale. We recommend instead using a data warehouse such as BigQuery,
23
+
Snowflake, Redshift. There is experimental support for Spark as well.
24
+
4. Setup CI/CD + dev vs staging vs prod environments to automatically update the registry as you change Feast feature definitions. See [docs](https://docs.feast.dev/how-to-guides/running-feast-in-production#1.-automatically-deploying-changes-to-your-feature-definitions).
25
+
5. (optional) Regularly scheduled materialization to power low latency feature retrieval (e.g. via Airflow). See [Batch data ingestion](https://docs.feast.dev/getting-started/concepts/data-ingestion#batch-data-ingestion)
26
+
for more details.
27
+
6. (optional) Deploy feature server instances with `feast serve` to expose endpoints to retrieve online features.
28
+
- See [Python feature server](https://docs.feast.dev/reference/feature-servers/python-feature-server) for details.
29
+
- Use cases can also directly call the Feast client to fetch features as per [Feature retrieval](https://docs.feast.dev/getting-started/concepts/feature-retrieval)
0 commit comments