Weaviate - 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 a Weaviate database. The requirements are as follows.

For the Unstructured UI or the Unstructured API: only Weaviate Cloud clusters are supported.
For Unstructured Ingest: Weaviate Cloud clusters, Weaviate installed locally, and Embedded Weaviate are supported.
For Weaviate installed locally, you will need the name of the target collection on the local instance.
For Embedded Weaviate, you will need the instance’s connection URL and the name of the target collection on the instance.
For Weaviate Cloud, you will need:
- A Weaviate database instance. The following information assumes that you have a Weaviate Cloud (WCD) account with a Weaviate database cluster in that account. Create a WCD account. Create a database cluster. For other database options, learn more.
- The URL and API key for the database cluster. Get the URL and API key.
- The name of the target collection in the database. Create a collection. An existing collection is not required. At runtime, the collection behavior is as follows: For the Unstructured UI or the Unstructured API:
  - If an existing collection name is specified, and Unstructured generates embeddings, but the number of dimensions that are generated does not match the existing collection’s embedding settings, the run will fail. You must change your Unstructured embedding settings or your existing collection’s embedding settings to match, and try the run again.
  - If a collection name is not specified, Unstructured creates a new collection in your Weaviate cluster. If Unstructured generates embeddings, the new collection’s name will be U<short-workflow-id>_<short-embedding-model-name>_<number-of-dimensions>. If Unstructured does not generate embeddings, the new collection’s name will be U<short-workflow-id.
  For Unstructured Ingest:
  - If an existing collection name is specified, and Unstructured generates embeddings, but the number of dimensions that are generated does not match the existing collection’s embedding settings, the run will fail. You must change your Unstructured embedding settings or your existing collection’s embedding settings to match, and try the run again.
  - If a collection name is not specified, Unstructured creates a new collection in your Weaviate cluster. The new collection’s name will be Unstructuredautocreated.
  If Unstructured creates a new collection and generates embeddings, you will not see an embeddings property in tools such as the Weaviate Cloud Collections user interface. To view the generated embeddings, you can run a Weaviate GraphQL query such as the following. In this query, replace <collection-name> with the name of the new collection, and replace <property-name> with the name of each additional available property that you want to return results for, such as text, type, element_id, record_id, and so on. The embeddings will be returned in the vector property.
  { Get { <collection-name> { _additional { vector } <property-name> <property-name> } } }

Weaviate requires an existing collection to have a data schema before you add data. At minimum, this schema must contain the record_id property, as follows:

{
    "class": "Elements",
    "properties": [
        {
            "name": "record_id",
            "dataType": ["text"]
        }
    ]
}

Weaviate generates any additional properties based on the incoming data. If you have specific schema requirements, you can define the schema manually. Unstructured cannot provide a schema that is guaranteed to work for everyone in all circumstances. This is because these schemas will vary based on your source files’ types; how you want Unstructured to partition, chunk, and generate embeddings; any custom post-processing code that you run; and other factors. You can adapt the following collection schema example for your own specific schema requirements:

{
    "class": "Elements",
    "properties": [
        {
            "name": "record_id",
            "dataType": ["text"]
        },
        {
            "name": "element_id",
            "dataType": ["text"]
        },
        {
            "name": "text",
            "dataType": ["text"]
        },
        {
            "name": "embeddings",
            "dataType": ["number[]"]
        },
        {
            "name": "metadata",
            "dataType": ["object"],
            "nestedProperties": [
                {
                    "name": "parent_id",
                    "dataType": ["text"]
                },
                {
                    "name": "page_number",
                    "dataType": ["text"]
                },
                {
                    "name": "is_continuation",
                    "dataType": ["boolean"]
                },
                {
                    "name": "orig_elements",
                    "dataType": ["text"]
                },
                {
                    "name": "partitioner_type",
                    "dataType": ["text"]
                }
            ]
        }
    ]
}

See also :

The Weaviate connector dependencies:

CLI, Python

pip install "unstructured-ingest[weaviate]"

You might also need to install additional dependencies, depending on your needs. Learn more. The following environment variables:

For Weaviate installed locally, WEAVIATE_COLLECTION - The name of the target collection in the instance, represented by --collection (CLI) or collection (Python).
For Embedded Weaviate:
- WEAVIATE_HOST - The connection URL to the instance, represented by --hostname (CLI) or hostname (Python).
- WEAVIATE_COLLECTION - The name of the target collection in the instance, represented by --collection (CLI) or collection (Python). If no value is provided, see the beginning of this article for the behavior at run time.
For Weaviate Cloud:
- WEAVIATE_CLUSTER_URL - THE REST endpoint for the Weaviate database cluster, represented by --cluster-url (CLI) or cluster_url (Python).
- WEAVIATE_API_KEY - The API key for the database cluster, represented by --api-key (CLI) or api_key (Python).
  For the CLI, the --api-key option here is part of the weaviate-cloud command. For Python, the api_key parameter here is part of the CloudWeaviateAccessConfig object.
- WEAVIATE_COLLECTION - The name of the target collection in the database, represented by --collection (CLI) or collection (Python). If no value is provided, see the beginning of this article for the behavior at run time.

Now call the Unstructured Ingest CLI or the Unstructured Ingest Python library. 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 is optional.

# For Weaviate installed locally:
unstructured-ingest \
  local \
    --input-path $LOCAL_FILE_INPUT_DIR \
    --partition-by-api \
    --api-key $UNSTRUCTURED_API_KEY \
    --partition-endpoint $UNSTRUCTURED_API_URL \
    --chunking-strategy by_title \
    --embedding-provider huggingface \
    --strategy hi_res \
    --additional-partition-args="{\"split_pdf_page\":\"true\", \"split_pdf_allow_failed\":\"true\", \"split_pdf_concurrency_level\": 15}" \
  weaviate-local \
    --collection $WEAVIATE_COLLECTION

# For Embedded Weaviate:
unstructured-ingest \
  local \
    --input-path $LOCAL_FILE_INPUT_DIR \
    --partition-by-api \
    --api-key $UNSTRUCTURED_API_KEY \
    --partition-endpoint $UNSTRUCTURED_API_URL \
    --chunking-strategy by_title \
    --embedding-provider huggingface \
    --strategy hi_res \
    --additional-partition-args="{\"split_pdf_page\":\"true\", \"split_pdf_allow_failed\":\"true\", \"split_pdf_concurrency_level\": 15}" \
  weaviate-embedded \
    --hostname $WEAVIATE_HOST \
    --collection $WEAVIATE_COLLECTION

# For Weaviate Cloud:
unstructured-ingest \
  local \
    --input-path $LOCAL_FILE_INPUT_DIR \
    --partition-by-api \
    --api-key $UNSTRUCTURED_API_KEY \
    --partition-endpoint $UNSTRUCTURED_API_URL \
    --chunking-strategy by_title \
    --embedding-provider huggingface \
    --strategy hi_res \
    --additional-partition-args="{\"split_pdf_page\":\"true\", \"split_pdf_allow_failed\":\"true\", \"split_pdf_concurrency_level\": 15}" \
  weaviate-cloud \
    --cluster-url $WEAVIATE_URL \
    --api-key $WEAVIATE_API_KEY \
    --collection $WEAVIATE_COLLECTION

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 self-hosted instance of the Unstructured API.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 self-hosted, 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