Editing article

Title

Summary

Content

Vertex AI Search is a fully-managed platform, powered by large language models, that lets you build AI-enabled search and recommendation experiences for your public or private websites or mobile applications. For a thorough understanding of the methodology and procedures associated with the creation, deployment, maintenance, and monitoring of a Vertex AI Search instance, please refer to the <a href="https://cloud.google.com/generative-ai-app-builder/docs/introduction">Vertex AI Search and Conversation documentation</a>.There are many options to DIY a Search/Chat bot … BUT Demos are easy, Production is hard.<figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*UeqxMw2saNXhowf7" /></figure>The following are the steps involved in constructing a Retrieval-Augmented Generation (RAG) search application:<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y3LZ8-03KQ_pN7Qy" /></figure>The utilization of managed Vertex AI Search by Google Cloud considerably streamlines this process.<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7sUEga74WmFPm_pr" /></figure>Follow the simple steps to build your datastores and engines, then load and query the data in GCP Vertex AI Search and Conversation platform.What we are going to Build<ol><li>Datastores and Engines from Terraform</li><li>Load data and query from python</li></ol><h3>Step 1:</h3><pre># Create Test DataStore resource &quot;google_discovery_engine_data_store&quot; &quot;test-ds&quot; { location = &quot;global&quot; data_store_id = &quot;test-data-store-id&quot; display_name = &quot;test-unstructured-datastore&quot; industry_vertical = &quot;GENERIC&quot; content_config = &quot;CONTENT_REQUIRED&quot; solution_types = [&quot;SOLUTION_TYPE_SEARCH&quot;] create_advanced_site_search = false project = &lt;PROJECT ID&gt; } # Create Test Serach Engine resource &quot;google_discovery_engine_search_engine&quot; &quot;test-engine&quot; { engine_id = &quot;test_engine_id&quot; collection_id = &quot;default_collection&quot; location = google_discovery_engine_data_store.test-ds.location display_name = &quot;test-engine&quot; industry_vertical = &quot;GENERIC&quot; data_store_ids = [google_discovery_engine_data_store.test-ds.data_store_id] common_config { company_name = &quot;Test Company&quot; } search_engine_config { search_add_ons = [&quot;SEARCH_ADD_ON_LLM&quot;] } project = &lt;PROJECT ID&gt; } # Output resource Id&#39;s output &quot;test_data_store_id&quot; { value =resource.google_discovery_engine_data_store.test-ds.data_store_id } output &quot;test_engine_id&quot; { value = resource.google_discovery_engine_search_engine.test-engine.engine_id }</pre>Run<pre>Plugin project ID in above terraform code and run below commands terraform init terraform plan terraform apply</pre>If you running terraform with your own GCP identity you will run into an error like below.<pre>Error: Error creating DataStore: googleapi: Error 403: Your application is authenticating by using local Application Default Credentials. The discoveryengine.googleapis.com API requires a quota project, which is not set by default. To learn how to set your quota project, see https://cloud.google.com/docs/authentication/adc-troubleshooting/user-creds .</pre>Add provider block to specify billing project. billing project and project where datastores are created can be same.<pre>provider &quot;google&quot; { project = &lt;PROJECT ID&gt; // Your actual GCP project ID user_project_override = true billing_project = &lt;BILLING PROJECT ID&gt; // The project used for quota }</pre>Once terraform is successful you can login to GCP Console -&gt; Vertex Search and Conversation<ol><li>Confirm an App named “test-engine” is created</li><li>Confirm an empty datastore named “test-unstructured-datastore” is created</li></ol><h3>Step 2:</h3>Create load_data.py with below contents. Plug in project Id and Datastore ID (output from terraform)<pre> from google.cloud import storage from google.api_core.client_options import ClientOptions from google.cloud import discoveryengine_v1alpha as discoveryengine def import_documents( project_id: str, location: str, data_store_id: str, gcs_uri: str, ): # Create a client client_options = ( ClientOptions( api_endpoint=f&quot;{location}-discoveryengine.googleapis.com&quot;) if location != &quot;global&quot; else None ) client = discoveryengine.DocumentServiceClient( client_options=client_options) # The full resource name of the search engine branch. # e.g. projects/{project}/locations/{location}/dataStores/{data_store_id}/branches/{branch} parent = client.branch_path( project=project_id, location=location, data_store=data_store_id, branch=&quot;default_branch&quot;, ) source_documents = [f&quot;{gcs_uri}/*&quot;] request = discoveryengine.ImportDocumentsRequest( parent=parent, gcs_source=discoveryengine.GcsSource( input_uris=source_documents, data_schema=&quot;content&quot; ), # Options: `FULL`, `INCREMENTAL` reconciliation_mode=discoveryengine.ImportDocumentsRequest.ReconciliationMode.INCREMENTAL, ) # Make the request operation = client.import_documents(request=request) response = operation.result() # Once the operation is complete, # get information from operation metadata metadata = discoveryengine.ImportDocumentsMetadata(operation.metadata) # Handle the response return operation.operation.name source_documents_gs_uri = ( &quot;gs://cloud-samples-data/gen-app-builder/search/alphabet-investor-pdfs&quot; ) PROJECT_ID = &lt;PROJECT ID&gt; DATASTORE_ID = &lt;DATASTORE ID&gt; LOCATION = &quot;global&quot; print(&quot; Starting loading data into datastore&quot;) import_documents(PROJECT_ID, LOCATION, DATASTORE_ID, source_documents_gs_uri) print(&quot; Completed loading data into datastore&quot;)</pre>Create search_data.py with below contents. Plug in project Id and Datastore ID (output from terraform)<pre>from google.cloud import storage from google.api_core.client_options import ClientOptions from google.cloud import discoveryengine_v1alpha as discoveryengine from typing import List def search_sample( project_id: str, location: str, data_store_id: str, search_query: str, ) -&gt; List[discoveryengine.SearchResponse]: # For more information, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/locations#specify_a_multi-region_for_your_data_store client_options = ( ClientOptions( api_endpoint=f&quot;{location}-discoveryengine.googleapis.com&quot;) if LOCATION != &quot;global&quot; else None ) # Create a client client = discoveryengine.SearchServiceClient(client_options=client_options) # The full resource name of the search engine serving config # e.g. projects/{project_id}/locations/{location}/dataStores/{data_store_id}/servingConfigs/{serving_config_id} serving_config = client.serving_config_path( project=project_id, location=location, data_store=data_store_id, serving_config=&quot;default_config&quot;, ) # Optional: Configuration options for search # Refer to the `ContentSearchSpec` reference for all supported fields: # https://cloud.google.com/python/docs/reference/discoveryengine/latest/google.cloud.discoveryengine_v1.types.SearchRequest.ContentSearchSpec content_search_spec = discoveryengine.SearchRequest.ContentSearchSpec( # For information about snippets, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/snippets snippet_spec=discoveryengine.SearchRequest.ContentSearchSpec.SnippetSpec( return_snippet=True ), # For information about search summaries, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/get-search-summaries summary_spec=discoveryengine.SearchRequest.ContentSearchSpec.SummarySpec( summary_result_count=5, include_citations=True, ignore_adversarial_query=True, ignore_non_summary_seeking_query=True, ), ) # Refer to the `SearchRequest` reference for all supported fields: # https://cloud.google.com/python/docs/reference/discoveryengine/latest/google.cloud.discoveryengine_v1.types.SearchRequest request = discoveryengine.SearchRequest( serving_config=serving_config, query=search_query, page_size=10, content_search_spec=content_search_spec, query_expansion_spec=discoveryengine.SearchRequest.QueryExpansionSpec( condition=discoveryengine.SearchRequest.QueryExpansionSpec.Condition.AUTO, ), spell_correction_spec=discoveryengine.SearchRequest.SpellCorrectionSpec( mode=discoveryengine.SearchRequest.SpellCorrectionSpec.Mode.AUTO ), ) response = client.search(request) return response QUERY = &quot;Who is the CEO of Google?&quot; PROJECT_ID = &lt;PROJECT ID&gt; DATASTORE_ID = &lt;DATASTORE ID&gt; LOCATION = &quot;global&quot; print(search_sample(PROJECT_ID, LOCATION, DATASTORE_ID, QUERY).summary.summary_text)</pre>Output of above python script should print the summary for the query.Do create datastores and engines in python refer to this <a href="https://github.com/GoogleCloudPlatform/generative-ai/blob/main/search/create_datastore_and_search.ipynb">Github Resource</a><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=364cdc591167" width="1" height="1" alt=""><hr><a href="https://medium.com/google-cloud/vertex-search-and-conversation-364cdc591167">Vertex Search and Conversation</a> was originally published in <a href="https://medium.com/google-cloud">Google Cloud - Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.

Author

Link

Published date

Image url

Feed url

Guid

Hidden blurb

--- !ruby/object:Feedjira::Parser::RSSEntry
title: Vertex Search and Conversation
url: https://medium.com/google-cloud/vertex-search-and-conversation-364cdc591167?source=rss----e52cf94d98af---4
author: Anita Gutta
categories:
- google-cloud-platform
- gcp-chat-bots
- conversational-ai
- machine-learning
- vertex-search
published: 2024-04-08 06:26:09.000000000 Z
entry_id: !ruby/object:Feedjira::Parser::GloballyUniqueIdentifier
 is_perma_link: 'false'
 guid: https://medium.com/p/364cdc591167
carlessian_info:
 news_filer_version: 2
 newspaper: Google Cloud - Medium
 macro_region: Blogs
rss_fields:
- title
- url
- author
- categories
- published
- entry_id
- content
content: 'Vertex AI Search is a fully-managed platform, powered by large language
 models, that lets you build AI-enabled search and recommendation experiences for
 your public or private websites or mobile applications. For a thorough understanding
 of the methodology and procedures associated with the creation, deployment, maintenance,
 and monitoring of a Vertex AI Search instance, please refer to the <a href="https://cloud.google.com/generative-ai-app-builder/docs/introduction">Vertex
 AI Search and Conversation documentation</a>.There are many options to DIY
 a Search/Chat bot … BUT Demos are easy, Production
 is hard.<figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*UeqxMw2saNXhowf7"
 /></figure>The following are the steps involved in constructing a Retrieval-Augmented
 Generation (RAG) search application:<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y3LZ8-03KQ_pN7Qy"
 /></figure>The utilization of managed Vertex AI Search by Google Cloud considerably
 streamlines this process.<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7sUEga74WmFPm_pr"
 /></figure>Follow the simple steps to build your datastores and engines, then
 load and query the data in GCP Vertex AI Search and Conversation platform.What
 we are going to Build<ol><li>Datastores and Engines from Terraform</li><li>Load
 data and query from python</li></ol><h3>Step 1:</h3><pre># Create
 Test DataStore resource &quot;google_discovery_engine_data_store&quot; &quot;test-ds&quot;
 { location = &quot;global&quot; data_store_id = &quot;test-data-store-id&quot; display_name
 = &quot;test-unstructured-datastore&quot; industry_vertical = &quot;GENERIC&quot; content_config
 = &quot;CONTENT_REQUIRED&quot; solution_types = [&quot;SOLUTION_TYPE_SEARCH&quot;] create_advanced_site_search
 = false project = &lt;PROJECT ID&gt; } # Create Test Serach Engine resource
 &quot;google_discovery_engine_search_engine&quot; &quot;test-engine&quot; { engine_id
 = &quot;test_engine_id&quot; collection_id = &quot;default_collection&quot; location
 = google_discovery_engine_data_store.test-ds.location display_name = &quot;test-engine&quot; industry_vertical
 = &quot;GENERIC&quot; data_store_ids = [google_discovery_engine_data_store.test-ds.data_store_id] common_config
 { company_name = &quot;Test Company&quot; } search_engine_config { search_add_ons
 = [&quot;SEARCH_ADD_ON_LLM&quot;] } project = &lt;PROJECT ID&gt; } #
 Output resource Id&#39;s output &quot;test_data_store_id&quot; { value =resource.google_discovery_engine_data_store.test-ds.data_store_id } output
 &quot;test_engine_id&quot; { value = resource.google_discovery_engine_search_engine.test-engine.engine_id }</pre>Run<pre>Plugin
 project ID in above terraform code and run below commands terraform init terraform
 plan terraform apply</pre>If you running terraform with your own GCP identity
 you will run into an error like below.<pre>Error: Error creating DataStore:
 googleapi: Error 403: Your application is authenticating by using local Application
 Default Credentials. The discoveryengine.googleapis.com API requires a quota
 project, which is not set by default. To learn how to set your quota project,
 see https://cloud.google.com/docs/authentication/adc-troubleshooting/user-creds
 .</pre>Add provider block to specify billing project. billing project and project
 where datastores are created can be same.<pre>provider &quot;google&quot; { project =
 &lt;PROJECT ID&gt; // Your actual GCP project ID user_project_override = true
 billing_project = &lt;BILLING PROJECT ID&gt; // The project used for quota }</pre>Once
 terraform is successful you can login to GCP Console -&gt; Vertex Search and Conversation<ol><li>Confirm
 an App named “test-engine” is created</li><li>Confirm an empty datastore named “test-unstructured-datastore”
 is created</li></ol><h3>Step 2:</h3>Create load_data.py with below contents.
 Plug in project Id and Datastore ID (output from terraform)<pre> from google.cloud
 import storage from google.api_core.client_options import ClientOptions from
 google.cloud import discoveryengine_v1alpha as discoveryengine def import_documents( project_id:
 str, location: str, data_store_id: str, gcs_uri: str, ): #
 Create a client client_options = ( ClientOptions( api_endpoint=f&quot;{location}-discoveryengine.googleapis.com&quot;) if
 location != &quot;global&quot; else None ) client = discoveryengine.DocumentServiceClient( client_options=client_options) #
 The full resource name of the search engine branch. # e.g. projects/{project}/locations/{location}/dataStores/{data_store_id}/branches/{branch} parent
 = client.branch_path( project=project_id, location=location, data_store=data_store_id, branch=&quot;default_branch&quot;, ) source_documents
 = [f&quot;{gcs_uri}/*&quot;] request = discoveryengine.ImportDocumentsRequest( parent=parent, gcs_source=discoveryengine.GcsSource( input_uris=source_documents,
 data_schema=&quot;content&quot; ), # Options: `FULL`, `INCREMENTAL` reconciliation_mode=discoveryengine.ImportDocumentsRequest.ReconciliationMode.INCREMENTAL, ) #
 Make the request operation = client.import_documents(request=request) response
 = operation.result() # Once the operation is complete, # get information
 from operation metadata metadata = discoveryengine.ImportDocumentsMetadata(operation.metadata) #
 Handle the response return operation.operation.name source_documents_gs_uri
 = ( &quot;gs://cloud-samples-data/gen-app-builder/search/alphabet-investor-pdfs&quot; ) PROJECT_ID
 = &lt;PROJECT ID&gt; DATASTORE_ID = &lt;DATASTORE ID&gt; LOCATION = &quot;global&quot; print(&quot;
 Starting loading data into datastore&quot;) import_documents(PROJECT_ID, LOCATION,
 DATASTORE_ID, source_documents_gs_uri) print(&quot; Completed loading data into
 datastore&quot;)</pre>Create search_data.py with below contents. Plug in project
 Id and Datastore ID (output from terraform)<pre>from google.cloud import storage from
 google.api_core.client_options import ClientOptions from google.cloud import
 discoveryengine_v1alpha as discoveryengine from typing import List def
 search_sample( project_id: str, location: str, data_store_id:
 str, search_query: str, ) -&gt; List[discoveryengine.SearchResponse]: # For
 more information, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/locations#specify_a_multi-region_for_your_data_store client_options
 = ( ClientOptions( api_endpoint=f&quot;{location}-discoveryengine.googleapis.com&quot;) if
 LOCATION != &quot;global&quot; else None ) # Create
 a client client = discoveryengine.SearchServiceClient(client_options=client_options) #
 The full resource name of the search engine serving config # e.g. projects/{project_id}/locations/{location}/dataStores/{data_store_id}/servingConfigs/{serving_config_id} serving_config
 = client.serving_config_path( project=project_id, location=location, data_store=data_store_id, serving_config=&quot;default_config&quot;, ) #
 Optional: Configuration options for search # Refer to the `ContentSearchSpec`
 reference for all supported fields: # https://cloud.google.com/python/docs/reference/discoveryengine/latest/google.cloud.discoveryengine_v1.types.SearchRequest.ContentSearchSpec content_search_spec
 = discoveryengine.SearchRequest.ContentSearchSpec( # For information
 about snippets, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/snippets snippet_spec=discoveryengine.SearchRequest.ContentSearchSpec.SnippetSpec( return_snippet=True ), #
 For information about search summaries, refer to: # https://cloud.google.com/generative-ai-app-builder/docs/get-search-summaries summary_spec=discoveryengine.SearchRequest.ContentSearchSpec.SummarySpec( summary_result_count=5, include_citations=True, ignore_adversarial_query=True, ignore_non_summary_seeking_query=True, ), ) #
 Refer to the `SearchRequest` reference for all supported fields: # https://cloud.google.com/python/docs/reference/discoveryengine/latest/google.cloud.discoveryengine_v1.types.SearchRequest request
 = discoveryengine.SearchRequest( serving_config=serving_config, query=search_query, page_size=10, content_search_spec=content_search_spec, query_expansion_spec=discoveryengine.SearchRequest.QueryExpansionSpec( condition=discoveryengine.SearchRequest.QueryExpansionSpec.Condition.AUTO, ), spell_correction_spec=discoveryengine.SearchRequest.SpellCorrectionSpec( mode=discoveryengine.SearchRequest.SpellCorrectionSpec.Mode.AUTO ), ) response
 = client.search(request) return response QUERY = &quot;Who is
 the CEO of Google?&quot; PROJECT_ID = &lt;PROJECT ID&gt; DATASTORE_ID
 = &lt;DATASTORE ID&gt; LOCATION = &quot;global&quot; print(search_sample(PROJECT_ID,
 LOCATION, DATASTORE_ID, QUERY).summary.summary_text)</pre>Output of above python
 script should print the summary for the query.Do create datastores and engines
 in python refer to this <a href="https://github.com/GoogleCloudPlatform/generative-ai/blob/main/search/create_datastore_and_search.ipynb">Github Resource</a><img
 src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=364cdc591167"
 width="1" height="1" alt=""><hr><a href="https://medium.com/google-cloud/vertex-search-and-conversation-364cdc591167">Vertex
 Search and Conversation</a> was originally published in <a href="https://medium.com/google-cloud">Google
 Cloud - Community</a> on Medium, where people are continuing the conversation by
 highlighting and responding to this story.'

Language

Active

Ricc internal notes

Ricc source

Show this article Back to articles