Skip to content

Commit 0c30e96

Browse files
lokeshranginenilokeshrangineni
authored
fix: Updating the instructions for quickstart guide. (#4120)
Updating the instructions for quickstart guide as per new instructions or output and also fixes the issue - 4104
1 parent 9184dde commit 0c30e96

1 file changed

Lines changed: 105 additions & 38 deletions

File tree

‎docs/getting-started/quickstart.md‎

Lines changed: 105 additions & 38 deletions
Original file line numberDiff line numberDiff line change
@@ -109,7 +109,7 @@ from feast import (
109109
from feast.on_demand_feature_view import on_demand_feature_view
110110
from feast.types import Float32, Float64, Int64
111111

112-
# Define an entity for the driver. You can think of entity as a primary key used to
112+
# Define an entity for the driver. You can think of an entity as a primary key used to
113113
# fetch features.
114114
driver = Entity(name="driver", join_keys=["driver_id"])
115115

@@ -138,7 +138,7 @@ driver_stats_fv = FeatureView(
138138
schema=[
139139
Field(name="conv_rate", dtype=Float32),
140140
Field(name="acc_rate", dtype=Float32),
141-
Field(name="avg_daily_trips", dtype=Int64),
141+
Field(name="avg_daily_trips", dtype=Int64, description="Average daily trips"),
142142
],
143143
online=True,
144144
source=driver_stats_source,
@@ -147,12 +147,6 @@ driver_stats_fv = FeatureView(
147147
tags={"team": "driver_performance"},
148148
)
149149

150-
# Defines a way to push data (to be available offline, online or both) into Feast.
151-
driver_stats_push_source = PushSource(
152-
name="driver_stats_push_source",
153-
batch_source=driver_stats_source,
154-
)
155-
156150
# Define a request data source which encodes features / information only
157151
# available at request time (e.g. part of the user initiated HTTP request)
158152
input_request = RequestSource(
@@ -191,6 +185,51 @@ driver_activity_v1 = FeatureService(
191185
driver_activity_v2 = FeatureService(
192186
name="driver_activity_v2", features=[driver_stats_fv, transformed_conv_rate]
193187
)
188+
189+
# Defines a way to push data (to be available offline, online or both) into Feast.
190+
driver_stats_push_source = PushSource(
191+
name="driver_stats_push_source",
192+
batch_source=driver_stats_source,
193+
)
194+
195+
# Defines a slightly modified version of the feature view from above, where the source
196+
# has been changed to the push source. This allows fresh features to be directly pushed
197+
# to the online store for this feature view.
198+
driver_stats_fresh_fv = FeatureView(
199+
name="driver_hourly_stats_fresh",
200+
entities=[driver],
201+
ttl=timedelta(days=1),
202+
schema=[
203+
Field(name="conv_rate", dtype=Float32),
204+
Field(name="acc_rate", dtype=Float32),
205+
Field(name="avg_daily_trips", dtype=Int64),
206+
],
207+
online=True,
208+
source=driver_stats_push_source, # Changed from above
209+
tags={"team": "driver_performance"},
210+
)
211+
212+
213+
# Define an on demand feature view which can generate new features based on
214+
# existing feature views and RequestSource features
215+
@on_demand_feature_view(
216+
sources=[driver_stats_fresh_fv, input_request], # relies on fresh version of FV
217+
schema=[
218+
Field(name="conv_rate_plus_val1", dtype=Float64),
219+
Field(name="conv_rate_plus_val2", dtype=Float64),
220+
],
221+
)
222+
def transformed_conv_rate_fresh(inputs: pd.DataFrame) -> pd.DataFrame:
223+
df = pd.DataFrame()
224+
df["conv_rate_plus_val1"] = inputs["conv_rate"] + inputs["val_to_add"]
225+
df["conv_rate_plus_val2"] = inputs["conv_rate"] + inputs["val_to_add_2"]
226+
return df
227+
228+
229+
driver_activity_v3 = FeatureService(
230+
name="driver_activity_v3",
231+
features=[driver_stats_fresh_fv, transformed_conv_rate_fresh],
232+
)
194233
```
195234
{% endtab %}
196235
{% endtabs %}
@@ -254,10 +293,14 @@ feast apply
254293
```
255294
Created entity driver
256295
Created feature view driver_hourly_stats
296+
Created feature view driver_hourly_stats_fresh
257297
Created on demand feature view transformed_conv_rate
298+
Created on demand feature view transformed_conv_rate_fresh
299+
Created feature service driver_activity_v3
258300
Created feature service driver_activity_v1
259301
Created feature service driver_activity_v2
260302
303+
Created sqlite table my_project_driver_hourly_stats_fresh
261304
Created sqlite table my_project_driver_hourly_stats
262305
```
263306
{% endtab %}
@@ -334,28 +377,40 @@ print(training_df.head())
334377
----- Feature schema -----
335378

336379
<class 'pandas.core.frame.DataFrame'>
337-
Int64Index: 3 entries, 0 to 2
338-
Data columns (total 6 columns):
339-
# Column Non-Null Count Dtype
340-
--- ------ -------------- -----
341-
0 event_timestamp 3 non-null datetime64[ns, UTC]
342-
1 driver_id 3 non-null int64
343-
2 label_driver_reported_satisfaction 3 non-null int64
344-
3 conv_rate 3 non-null float32
345-
4 acc_rate 3 non-null float32
346-
5 avg_daily_trips 3 non-null int32
347-
dtypes: datetime64[ns, UTC](1), float32(2), int32(1), int64(2)
348-
memory usage: 132.0 bytes
380+
RangeIndex: 3 entries, 0 to 2
381+
Data columns (total 10 columns):
382+
# Column Non-Null Count Dtype
383+
--- ------ -------------- -----
384+
0 driver_id 3 non-null int64
385+
1 event_timestamp 3 non-null datetime64[ns, UTC]
386+
2 label_driver_reported_satisfaction 3 non-null int64
387+
3 val_to_add 3 non-null int64
388+
4 val_to_add_2 3 non-null int64
389+
5 conv_rate 3 non-null float32
390+
6 acc_rate 3 non-null float32
391+
7 avg_daily_trips 3 non-null int32
392+
8 conv_rate_plus_val1 3 non-null float64
393+
9 conv_rate_plus_val2 3 non-null float64
394+
dtypes: datetime64[ns, UTC](1), float32(2), float64(2), int32(1), int64(4)
395+
memory usage: 336.0 bytes
349396
None
350397

351398
----- Example features -----
352399

353-
event_timestamp driver_id ... acc_rate avg_daily_trips
354-
0 2021-08-23 15:12:55.489091+00:00 1003 ... 0.077863 741
355-
1 2021-08-23 15:49:55.489089+00:00 1002 ... 0.074327 113
356-
2 2021-08-23 16:14:55.489075+00:00 1001 ... 0.105046 347
400+
driver_id event_timestamp label_driver_reported_satisfaction \
401+
0 1001 2021-04-12 10:59:42+00:00 1
402+
1 1002 2021-04-12 08:12:10+00:00 5
403+
2 1003 2021-04-12 16:40:26+00:00 3
404+
405+
val_to_add val_to_add_2 conv_rate acc_rate avg_daily_trips \
406+
0 1 10 0.800648 0.265174 643
407+
1 2 20 0.644141 0.996602 765
408+
2 3 30 0.855432 0.546345 954
357409

358-
[3 rows x 6 columns]
410+
conv_rate_plus_val1 conv_rate_plus_val2
411+
0 1.800648 10.800648
412+
1 2.644141 20.644141
413+
2 3.855432 30.855432
359414
```
360415
{% endtab %}
361416
{% endtabs %}
@@ -389,10 +444,20 @@ print(training_df.head())
389444
```
390445
----- Example features -----
391446
392-
driver_id event_timestamp ... acc_rate avg_daily_trips conv_rate_plus_val1
393-
0 1001 2022-08-08 18:22:06.555018+00:00 ... 0.864639 359 1.663844
394-
1 1002 2022-08-08 18:22:06.555018+00:00 ... 0.695982 311 2.151189
395-
2 1003 2022-08-08 18:22:06.555018+00:00 ... 0.949191 789 3.769165
447+
driver_id event_timestamp \
448+
0 1001 2024-04-19 14:58:16.452895+00:00
449+
1 1002 2024-04-19 14:58:16.452895+00:00
450+
2 1003 2024-04-19 14:58:16.452895+00:00
451+
452+
label_driver_reported_satisfaction val_to_add val_to_add_2 conv_rate \
453+
0 1 1 10 0.535773
454+
1 5 2 20 0.171976
455+
2 3 3 30 0.275669
456+
457+
acc_rate avg_daily_trips conv_rate_plus_val1 conv_rate_plus_val2
458+
0 0.689705 428 1.535773 10.535773
459+
1 0.737113 369 2.171976 20.171976
460+
2 0.156630 116 3.275669 30.275669
396461
```
397462
{% endtab %}
398463
{% endtabs %}
@@ -413,11 +478,13 @@ feast materialize-incremental $CURRENT_TIME
413478
{% tabs %}
414479
{% tab title="Output" %}
415480
```bash
416-
Materializing 1 feature views to 2021-08-23 16:25:46+00:00 into the sqlite online
417-
store.
481+
Materializing 2 feature views to 2024-04-19 10:59:58-04:00 into the sqlite online store.
418482

419-
driver_hourly_stats from 2021-08-22 16:25:47+00:00 to 2021-08-23 16:25:46+00:00:
420-
100%|████████████████████████████████████████████| 5/5 [00:00<00:00, 592.05it/s]
483+
driver_hourly_stats from 2024-04-18 15:00:46-04:00 to 2024-04-19 10:59:58-04:00:
484+
100%|████████████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 370.32it/s]
485+
driver_hourly_stats_fresh from 2024-04-18 15:00:46-04:00 to 2024-04-19 10:59:58-04:00:
486+
100%|███████████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 1046.64it/s]
487+
Materializing 2 feature views to 2024-04-19 10:59:58-04:00 into the sqlite online store.
421488
```
422489
{% endtab %}
423490
{% endtabs %}
@@ -458,11 +525,11 @@ pprint(feature_vector)
458525
{% tab title="Output" %}
459526
```bash
460527
{
461-
'acc_rate': [0.5732735991477966, 0.7828438878059387],
462-
'avg_daily_trips': [33, 984],
463-
'conv_rate': [0.15498852729797363, 0.6263588070869446],
528+
'acc_rate': [0.25351759791374207, 0.8949751853942871],
529+
'avg_daily_trips': [712, 791],
530+
'conv_rate': [0.5038306713104248, 0.9839504361152649],
464531
'driver_id': [1004, 1005]
465-
}
532+
}
466533
```
467534
{% endtab %}
468535
{% endtabs %}
@@ -479,7 +546,7 @@ The `driver_activity_v1` feature service pulls all features from the `driver_hou
479546
```python
480547
from feast import FeatureService
481548
driver_stats_fs = FeatureService(
482-
name="driver_activity_v1", features=[driver_hourly_stats_view]
549+
name="driver_activity_v1", features=[driver_stats_fv]
483550
)
484551
```
485552

0 commit comments

Comments
 (0)