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
-[User flow 1: ML Platform Team](#user-flow-1-ml-platform-team)
15
15
-[Step 0: Setup S3 bucket for registry and file sources](#step-0-setup-s3-bucket-for-registry-and-file-sources)
@@ -31,6 +31,14 @@ We focus on a specific example (that does not include online features + models):
31
31
-[User flow 3: Data Scientists](#user-flow-3-data-scientists)
32
32
-[Conclusion](#conclusion)
33
33
34
+
# Installing Feast
35
+
Before we get started, first install Feast with AWS dependencies. Due to a bug in Feast 0.21, we'll also need s3fs for this tutorial to directly fetch from an S3 source:
36
+
37
+
```bash
38
+
pip install "feast[aws]"
39
+
pip install s3fs
40
+
```
41
+
34
42
# Reviewing Feast concepts
35
43
Let's quickly review some Feast concepts needed to build this use case. You'll need:
36
44
| Concept | Requirements |
@@ -41,14 +49,6 @@ Let's quickly review some Feast concepts needed to build this use case. You'll n
41
49
| Registry | In `feature_store.yaml`, specifying a path (within an existing S3 bucket) the registry is written to. Users + model servers will pull from this to get the latest registered features + metadata |
42
50
| Transformations | Feast supports last mile transformations with `OnDemandFeatureView`s that can be re-used |
43
51
44
-
# Installing Feast
45
-
Before we get started, first install Feast with AWS dependencies. Due to a bug in Feast 0.21, we'll also need s3fs for this tutorial to directly fetch from an S3 source:
46
-
47
-
```bash
48
-
pip install "feast[aws]"
49
-
pip install s3fs
50
-
```
51
-
52
52
# User flows
53
53
There are three user groups here worth considering. The ML platform team, the data scientists, and the ML engineers scheduling models in batch.
0 commit comments