IBM watsonx.data - Unstructured

This page was recently updated. What do you think about it? Let us know!.

Batch process all your records to store structured outputs in IBM watsonx.data.

The requirements are as follows.

An IBM Cloud account. Create an IBM Cloud account if you do not already have one.
An API key for the IBM Cloud account. If you do not have one already, create one as follows:
1. Log in to your IBM Cloud account.
2. In the top navigation bar, click Manage and then, under Security and access, click Access (IAM).
3. On the sidebar, under Manage identities, click API keys.
4. With the View list showing My IBM Cloud API keys, click Create.
5. Enter some Name and an optional Description for the API key.
6. Leave Leaked action set to Disable the leaked key and Session creation set to No.
7. Click Create.
8. Click Copy or Download to copy or save the API key to a secure location. You won’t be able to access this API key from this screen again.
An IBM Cloud Object Storage (COS) instance in the account, and a bucket within that instance. If you do not have them already, create them as follows:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. Click Create resource.
4. With IBM Cloud catalog selected, search for and select Object Storage.
5. Complete the on-screen instructions to finish creating the COS instance.
6. With the COS instance’s settings page shown, on the Buckets tab, click Create bucket.
7. Complete the on-screen instructions to finish creating the bucket.
The name, region, and public endpoint for the target bucket within the target Cloud Object Storage (COS) instance. To get these:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Storage, and then click the target COS instance.
4. On the Buckets tab, click the target bucket.
5. On the Configuration tab, note the following:
  - Under Bucket details, note the Bucket name. This is the bucket’s name.
  - Under Bucket details section, note the value inside of the parentheses inside Location, for example us-east. This is the bucket’s region.
  - Under Endpoints, note the value of Public, for example s3.us-east.cloud-object-storage.appdomain.cloud. (Ignore the values of Private and Direct). This is the bucket’s public endpoint.
An HMAC access key ID and secret access key for the target Cloud Object Storage (COS) instance. If you do not have them already, get or create them as follows:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Storage, and then click the target COS instance.
4. On the Service credentials tab, if there is a credential that you want to use in the list, expand the credential, and copy the following values to a secure location:
  - access_key_id under cos_hmac_keys, which represents the HMAC access key ID.
  - secret_access_key under cos_hmac_keys, which represents the HMAC secret access key.
  After you have copied the preceding values, you have completed this procedure.
5. If there is not a credential that you want to use, or there are no credentials at all, click New Credential.
6. Enter some Name for the credential.
7. For Role, select at least Writer, leave Select Service ID set to Auto Generated, switch on Include HMAC Credential, and then click Add.
8. In the list of credentials, expand the credential, and copy the following values to a secure location:
  - access_key_id under cos_hmac_keys, which represents the HMAC access key ID.
  - secret_access_key under cos_hmac_keys, which represents the HMAC secret access key.
An IBM watsonx.data data store instance in the IBM Cloud account. If you do not have one already, create one as follows:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. Click Create resource.
4. With IBM Cloud catalog selected, search for and select watsonx.data.
5. Complete the on-screen instructions to finish creating the watsonx.data data store instance.
An Apache Iceberg-based catalog within the watsonx.data data store instance. If you do not have one already, create one as follows:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Databases, and then click the target watsonx.data data store instance.
4. Click Open web console.
5. If prompted, log in to the web console.
6. On the sidebar, click Infrastructure manager. If the sidebar is not visible, click the Global navigation icon to the far left of the top navigation bar.
7. Click Add component.
8. Under Storage, click IBM Cloud Object Storage, and then click Next.
9. Complete the on-screen instructions to finish creating the Iceberg catalog. This includes providing the following settings:
  - Some display name for the component.
  - The name of the target bucket within the target Cloud Object Storage (COS) instance that you noted earlier.
  - The region for the target bucket, which you noted earlier.
  - The public endpoint for the target bucket, which you noted earlier. For this screen only, be sure to prefix the public endpoint with https://.
  - The HMAC access key ID for the target COS instance, which you noted earlier.
  - The HMAC secret access key for the target COS instance, which you noted earlier.
10. Next to Connection status, click Test connection to test the connection. Do not proceed until Successful is shown. If the connection is not successful, check the values you entered for the target bucket name, region, endpoint, access key, and secret access key, and try again.
11. Check the box labelled Associate Catalog.
12. Check the box labelled Activate now.
13. Under Associated catalog, for Catalog type, select Apache Iceberg.
14. Enter some Catalog name.
15. Click Associate.
16. On the sidebar, click Infrastructure manager. Make sure the catalog is associated with the appropriate engines. If it is not, rest your mouse on an unassociated target engine, click the Manage associations icon, check the box next to the target catalog’s name, and then click Save and restart engine.
  
  To create an engine if one is not already shown, click Add component, and follow the on-screen to add an appropriate engine from the list of available Engines (for example, an IBM Presto engine).
The catalog name and metastore REST endpoint for the target Iceberg catalog. To get this:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Databases, and then click the target watsonx.data data store instance.
4. Click Open web console.
5. If prompted, log in to the web console.
6. On the sidebar, click Infrastructure manager. If the sidebar is not visible, click the Global navigation icon to the far left of the top navigation bar.
7. In the Catalogs section, click the target Iceberg catalog.
8. On the Details tab, note the value of Name representing the catalog name, and Metastore REST endpoint representing the metastore REST endpoint. (Ignore the Metastore Thrift endpoint value.)
A namespace (also known as a schema) and a table in the target catalog. If you do not have these already, create them as follows:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Databases, and then click the target watsonx.data data store instance.
4. Click Open web console.
5. If prompted, log in to the web console.
6. On the sidebar, click Data manager. If the sidebar is not visible, click the Global navigation icon to the far left of the top navigation bar.
7. On the Browse data tab, under Catalogs associated, click the target catalog.
8. Click the ellipses, and then click Create schema.
9. Enter some Name for the schema, and then click Create.
10. On the sidebar, click Query workspace.
11. In the SQL editor, enter and run a table creation statement such as the following, replacing <catalog-name> with the name of the target catalog and <schema-name> with the name of the target schema:
  CREATE TABLE <catalog-name>.<schema-name>.elements ( "type" varchar, "element_id" varchar, "text" varchar, "file_directory" varchar, "filename" varchar, "languages" array(varchar), "last_modified" double, "page_number" varchar, "filetype" varchar, "url" varchar, "version" varchar, "record_locator" varchar, "date_created" double, "date_modified" double, "date_processed" double, "filesize_bytes" bigint, "points" varchar, "system" varchar, "layout_width" bigint, "layout_height" bigint, "id" varchar, "record_id" varchar, "parent_id" varchar ) WITH ( delete_mode = 'copy-on-write', format = 'PARQUET', format_version = '2' )
  Note that incoming elements that do not have matching column names will be dropped upon record insertion. For example, if the incoming data has an element named sent_from and there is no column named sent_from in the table, the sent_from element will be dropped upon record insertion. You should modify the preceding sample table creation statement to add columns for any additional elements that you want to be included upon record insertion.
The name of the target namespace (also known as a schema) within the target catalog, and name of the target table within that schema. To get these:
1. Log in to your IBM Cloud account.
2. On the sidebar, click the Resource list icon. If the sidebar is not visible, click the Navigation Menu icon to the far left of the top navigation bar.
3. In the list of resources, expand Databases, and then click the target watsonx.data data store instance.
4. Click Open web console.
5. If prompted, log in to the web console.
6. On the sidebar, click Data manager. If the sidebar is not visible, click the Global navigation icon to the far left of the top navigation bar.
7. On the Browse data tab, expand the name of the target catalog, and note the names of the target schema and target table.
The name of the column in the target table that uniquely identifies each of the records in the table.

The IBM watsonx.data connector dependencies:

CLI, Python
pip install "unstructured-ingest[ibm-watsonx-s3]"

You might also need to install additional dependencies, depending on your needs. Learn more.

The following environment variables:

IBM_IAM_API_KEY - An API key for the target IBM Cloud account, represented by --iam-api-key (CLI) or iam_api_key (Python).
IBM_COS_ACCESS_KEY - An HMAC access key ID for the target IBM Cloud Object Storage (COS) instance, represented by --access-key-id (CLI) or access_key_id (Python).
IBM_COS_SECRET_ACCESS_KEY - The associated HMAC secret access key ID for the target HMAC access key, represented by --secret-access-key (CLI) or secret_access_key (Python).
IBM_ICEBERG_CATALOG_METASTORE_REST_ENDPOINT - The metastore REST endpoint value for the target Apache Iceberg catalog in the target IBM watsonx.data data store instance, represented by --iceberg_endpoint (CLI) or iceberg_endpoint (Python). Do not include https:// in this value.
IBM_COS_BUCKET_PUBLIC_ENDPOINT - The target COS instance’s endpoint value, represented by --object-storage-endpoint (CLI) or object_storage_endpoint (Python).
IBM_COS_BUCKET_REGION - The target COS instance’s region short ID, represented by --object-storage-region (CLI) or object_storage_region (Python).
IBM_ICEBERG_CATALOG - The name of the target Iceberg catalog, represented by --catalog (CLI) or catalog (Python).
IBM_ICEBERG_SCHEMA - The name of the target namespace (also known as a schema) in the target catalog, represented by --namespace (CLI) or namespace (Python).
IBM_ICEBERG_TABLE - The name of the target table in the target schema, represented by --table (CLI) or table (Python).
IBM_ICEBERG_TABLE_UNIQUE_RECORD_COLUMN - The name of the column that uniquely identifies each record in the target table, represented by --record-id-key (CLI) or record_id_key (Python). The default is record_id.

Additionally, --max-retries (CLI) or max_retries (Python) is an optional parameter that specifies the number of times to retry uploading data. The default is 5. If specified, it must be a number between 2 and 10, inclusive.

Now call the Unstructured Ingest CLI or Unstructured Ingest Python. The source connector can be any of the ones supported. This example uses the local source connector.

This example sends files to Unstructured for processing by default. To process files locally instead, see the instructions at the end of this page.

#!/usr/bin/env bash

# Chunking and embedding are optional.

unstructured-ingest \
  local \
    --input-path $LOCAL_FILE_INPUT_DIR \
    --chunk-elements \
    --embedding-provider huggingface \
    --partition-by-api \
    --api-key $UNSTRUCTURED_API_KEY\
    --partition-endpoint $UNSTRUCTURED_API_URL \
    --strategy hi_res \
    --additional-partition-args="{\"split_pdf_page\":\"true\", \"split_pdf_allow_failed\":\"true\", \"split_pdf_concurrency_level\": 15}" \
  ibm-watsonx-s3 \
    --iam-api-key $IBM_IAM_API_KEY \
    --access-key-id $IBM_COS_ACCESS_KEY \
    --secret-access-key $IBM_COS_SECRET_ACCESS_KEY \
    --iceberg-endpoint $IBM_ICEBERG_CATALOG_METASTORE_REST_ENDPOINT \
    --object-storage-endpoint $IBM_COS_BUCKET_PUBLIC_ENDPOINT \
    --object-storage-region $IBM_COS_BUCKET_REGION \
    --catalog $IBM_ICEBERG_CATALOG \
    --namespace $IBM_ICEBERG_SCHEMA \
    --table $IBM_ICEBERG_TABLE \
    --max-retries 5 \
    --record-id-key $IBM_ICEBERG_TABLE_UNIQUE_RECORD_COLUMN

For the Unstructured Ingest CLI and the Unstructured Ingest Python library, you can use the --partition-by-api option (CLI) or partition_by_api (Python) parameter to specify where files are processed:

To do local file processing, omit --partition-by-api (CLI) or partition_by_api (Python), or explicitly specify partition_by_api=False (Python).

Local file processing does not use an Unstructured API key or API URL, so you can also omit the following, if they appear:
- --api-key $UNSTRUCTURED_API_KEY (CLI) or api_key=os.getenv("UNSTRUCTURED_API_KEY") (Python)
- --partition-endpoint $UNSTRUCTURED_API_URL (CLI) or partition_endpoint=os.getenv("UNSTRUCTURED_API_URL") (Python)
- The environment variables UNSTRUCTURED_API_KEY and UNSTRUCTURED_API_URL
To send files to the Unstructured Partition Endpoint for processing, specify --partition-by-api (CLI) or partition_by_api=True (Python).

Unstructured also requires an Unstructured API key and API URL, by adding the following:
- --api-key $UNSTRUCTURED_API_KEY (CLI) or api_key=os.getenv("UNSTRUCTURED_API_KEY") (Python)
- --partition-endpoint $UNSTRUCTURED_API_URL (CLI) or partition_endpoint=os.getenv("UNSTRUCTURED_API_URL") (Python)
- The environment variables UNSTRUCTURED_API_KEY and UNSTRUCTURED_API_URL, representing your API key and API URL, respectively.
You must specify the API URL only if you are not using the default API URL for Unstructured Ingest, for example, if you are using a version of the Unstructured API that is hosted on your own compute infrastructure.
The default API URL for Unstructured Ingest is https://api.unstructuredapp.io/general/v0/general, which is the API URL for the Unstructured Partition Endpoint.
If you do not have an API key, get one now.
If the Unstructured API is hosted on your own compute infrastructure, the process for generating Unstructured API keys, and the Unstructured API URL that you use, are different. For details, contact Unstructured Sales at sales@unstructured.io.

Ingestion