Unstructured provides tools to extract content from files and transform it into clean documents ready to be chunked and embedded. For a list of supported formats, see Unstructured documentation. You can use free Unstructured API or paid Unstructured Serverless API.

Prerequisites

You need an API key to your Unstructured account.

Use Unstructured

First, connect deepset Cloud to Unstructured through the Connections page:

Click your initials in the top right corner and select Connections.
Click Connect next to the provider.
Enter your user access token and submit it.

Then, add the UnstructuredFileConverter component to your indexing pipeline.

Usage Examples

This is an example of an indexing pipeline that uses Unstructured API to process files:

components:
  ...
  unstructured_converter:
    type: haystack_integrations.components.converters.unstructured.converter.UnstructuredFileConverter
    init_parameters: {}

  splitter:
    type: deepset_cloud_custom_nodes.preprocessors.document_splitter.DeepsetDocumentSplitter
    init_parameters:
      split_by: word
      split_length: 250
      split_overlap: 30
      respect_sentence_boundary: True
      language: en

  document_embedder:
    type: haystack.components.embedders.sentence_transformers_document_embedder.SentenceTransformersDocumentEmbedder
    init_parameters:
      model: "intfloat/e5-base-v2"

  writer:
    type: haystack.components.writers.document_writer.DocumentWriter
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          embedding_dim: 768
          similarity: cosine
      policy: OVERWRITE

connections:  # Defines how the components are connected
- sender: unstructured_converter.documents
  receiver: splitter.documents
- sender: splitter.documents
  receiver: document_embedder.documents
- sender: document_embedder.documents
  receiver: writer.documents

max_loops_allowed: 100

inputs:  # Define the inputs for your pipeline
  files: "file_classifier.sources"  # This component will receive the files to index as input