Skip to content

Commit 60b1566

Browse files
authored
Update BedrockTextToSql_for_Athena.ipynb
Added a note about readme for clarification
1 parent 8d3e6a2 commit 60b1566

File tree

1 file changed

+1
-16
lines changed

1 file changed

+1
-16
lines changed

‎BedrockTextToSql_for_Athena.ipynb

Lines changed: 1 addition & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -12,9 +12,6 @@
1212
"source": [
1313
"# Building Text-to-SQL capability to Amazon Athena using Amazon Bedrock\n",
1414
"\n",
15-
"- **Use of amazon.titan-embed-text-v1 for creating embedding**\n",
16-
"- **Use of Amazon OpenSearch as a vector database**\n",
17-
"- **Use of anthropic.claude-v2:1 as base LLM Model**\n",
1815
"\n",
1916
"\n",
2017
"\n",
@@ -88,19 +85,7 @@
8885
"- An Amazon OpenSearch cluster for storing embeddings.Here Opensearch credenitals are in notebooks. However Opensearch cluster's access credentials (username and password) can be stored in AWS Secrets Mananger by following steps described [here](https://docs.aws.amazon.com/secretsmanager/latest/userguide/managing-secrets.html).\n",
8986
"\n",
9087
"**The workflow for this notebook is as follows:**\n",
91-
"1. Create an S3 bucket with the name \"knowledgebase-<ACCOUNT_ID>\" \n",
92-
" - create a folder \"input\" in that bucket\n",
93-
"2. Download data from source \n",
94-
" - https://developer.imdb.com/non-commercial-datasets/#titleakastsvgz and upload to S3 bucket from step 1 and into the \"input\" folder\n",
95-
" - https://developer.imdb.com/non-commercial-datasets/#titlebasicstsvgz and upload to S3 bucket from step 1 and into the \"input\" folder\n",
96-
"3. Glue Steps\n",
97-
" - Create a glue database \"imdb_stg\" \n",
98-
" - Create a glue crawler \"text-2-sql-crawler\" with the datasource set to the S3 bucket created in step 1. Run the crawler.\n",
99-
" - 2 tables should be created in Glue data catalo.g Make sure you are able to query through athena. \n",
100-
"4. From the Bedrock console, Create a new knowledgebase \n",
101-
"1. Install the required Python packages \n",
102-
"1. Create embedding and vector store. Do a similarity search with embeddings stored in the OpenSearch index for an input query.\n",
103-
"1. Execute this notebook to generate sql.."
88+
"*Please read [Readme.md](https://github.com/aws-samples/text-to-sql-for-athena/blob/claude3branch/README.md) to learn about the detailed steps*"
10489
]
10590
},
10691
{

0 commit comments

Comments
 (0)