Python scripts and Databricks notebooks to display and list some Databricks objects.
Last updated: _2024-05-20.
- get-model-serving-endpoint - get model serving endpoint
- list-model-serving-endpoints - list model serving endpoints
- list-vector-search-endpoints - list vector search endpoints
- get-vector-search-endpoint
- list-feature-tables - list feature tables (non-UC "Feature Store" instead of UC "Feature Engineering")
Get Databricks model serving endpoint details.
get-model-serving-endpoint \
--endpoint sklearn_wine_best \
--openapi True \
--output-file sklearn_wine_best.json
{
"name": "sklearn_wine_best",
"creator": "andre@my-company.com",
"creation_timestamp": 1706467706000,
"last_updated_timestamp": 1706467706000,
"state": {
"ready": "READY",
"config_update": "NOT_UPDATING"
},
"config": {
"served_entities": [
{
"name": "Sklearn_Wine_best-2",
"entity_name": "Sklearn_Wine_best",
"entity_version": "2",
"workload_size": "Small",
"scale_to_zero_enabled": true,
"workload_type": "CPU",
"state": {
"deployment": "DEPLOYMENT_READY",
"deployment_state_message": "Scaled to zero"
},
"creator": "andre@my-company.com",
"creation_timestamp": 1706467706000,
"_creation_timestamp": "2024-01-28 18:48:26"
}
],
"traffic_config": {
"routes": [
{
"served_model_name": "Sklearn_Wine_best-2",
get-model-serving-endpoint --help
Options:
--endpoint TEXT Model serving endpoint name.
--openapi BOOLEAN Get endpoint OpenAPI schema and append it to
the output JSON.
--use-deployment-client BOOLEAN
Databricks only. Use
DatabricksDeploymentClient. Otherwise, make
direct calls to 'api/2.0/serving-endpoints'.
--get-raw BOOLEAN Preserve raw JSON as received from API call.
[default: False]
--output-file TEXT JSON output file.
--silent BOOLEAN Do not display to stdout. [default: False]
--expand-model-version [none|version|version-and-signature|version-all]
Model type: version|version-and-
signature|version-all
List Databricks model serving endpoints.
list-model-serving-endpoints \
--columns name,endpoint_type,task,creator,creation_timestamp \
--output-file-base endpoints
| name | endpoint_type | task | creator | creation_timestamp |
|----------------------------------|----------------------|--------------------|-------------------------|----------------------|
| embeddings_proxy_eo | | | amelia@mycompany.com | 2023-11-22 18:36:33 |
| eo_rfp_rag | | | amelia@mycompany.com | 2023-11-22 19:00:33 |
| databricks-llama-2-70b-chat | FOUNDATION_MODEL_API | llm/v1/chat | | 2023-11-10 09:53:20 |
| databricks-mixtral-8x7b-instruct | FOUNDATION_MODEL_API | llm/v1/chat | | 2023-11-10 09:53:20 |
| databricks-bge-large-en | FOUNDATION_MODEL_API | llm/v1/embeddings | | 2023-11-10 09:53:20 |
| databricks-mpt-30b-instruct | FOUNDATION_MODEL_API | llm/v1/completions | | 2023-11-10 09:53:20 |
| databricks-mpt-7b-instruct | FOUNDATION_MODEL_API | llm/v1/completions | | 2023-11-10 09:53:20 |
| openai-completions-endpoint | EXERNAL_MODEL | llm/v1/completions | k1.denali@mycompany.com | 2024-05-10 19:03:00 |
Options:
--model-type [all|custom|foundation|external]
Model type: all|custom|foundation|external
--openapi BOOLEAN Write OpenAPI schema for all endpoints to
file '{output-file-base}_opena[i.json'.
--columns TEXT Columns to display. Comma delimited.
--output-file-base TEXT File base for JSON and CSV output files. For
example, 'out' will result in 'out.csv' and
'out.json.' [default: out]
--use-deployment-client BOOLEAN
Databricks only. Use
DatabricksDeploymentClient. Otherwise, make
direct calls to 'api/2.0/serving-endpoints'.
--get-raw BOOLEAN Preserve raw JSON as received from API call.
[default: False]
--get-details BOOLEAN Get details of each listed object.
--normalize-pandas-df BOOLEAN Convert JSON list with pd.json_normalize,
otherwise use pd.DataFrame.
List feature tables (non-UC "Feature Store" instead of UC "Feature Engineering").
list-feature-tables \
--columns name,primary_keys,features,creation_timestamp,creator_id
| name | primary_keys | features | creation_timestamp | creator_id |
|-------------------------------------------------------------------|------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------|-----------------------------|
| andre_fs_wine.wine_features | ['wine_id'] | ['alcohol', 'chlorides', 'citric_acid', 'density', 'fixed_acidity', 'free_sulfur_dioxide', 'pH', 'residual_sugar', 'sulphates', 'total_sulfur_dioxide', 'volatile_acidity', 'wine_id'] | 2023-11-16 11:22:54 | amelia.singer@mycompany.com |
| andre_fs_wine.wine_static_features | ['wine_id'] | ['chlorides', 'citric_acid', 'density', 'fixed_acidity', 'free_sulfur_dioxide', 'pH', 'residual_sugar', 'sulphates', 'total_sulfur_dioxide', 'volatile_acidity', 'wine_id'] | 2023-05-29 02:49:42 | amelia.singer@mycompany.com |
| dbdemos_fs_travel_shared.destination_features_advanced | ['destination_id'] | ['destination_id', 'sum_clicks_7d', 'sum_impressions_7d', 'ts'] | 2023-08-14 16:58:07 | amelia.singer@mycompany.com |
| travel_recommendations_realtime.destination_availability_features | ['destination_id', 'booking_date'] | ['availability', 'booking_date', 'destination_id', 'event_ts', 'name', 'price'] | 2022-10-31 18:42:57 | samaim.amisam@mycompany.com |
| travel_recommendations_realtime.destination_location_features | ['destination_id'] | ['destination_id', 'latitude', 'longitude', 'name'] | 2022-10-31 18:18:19 | samaim.amisam@mycompany.com |
| travel_recommendations_realtime.destination_popularity_features | ['destination_id'] | ['destination_id', 'sum_clicks_7d', 'sum_impressions_7d', 'ts'] | 2022-10-31 20:09:12 | samaim.amisam@mycompany.com |
Options:
--columns TEXT Columns to display. Comma delimited.
--output-csv-file TEXT Output CSV file.