Skip to content
API Reference
/
ai-index
/
ai_index_orgs_users_documents_graph_train_create

ibl-data-manager (3.59.0-ai-plus)

API for iblai

Download OpenAPI description
ibl.json
ibl.yaml
Languages
Servers
Mock server

https://docs.ibl.ai/_mock/apis/ibl/

https://base.manager.iblai.app/

ai-account

Operations

ai-analytics

Operations

ai-bot

Operations

ai-finetuning

Operations

ai-index

Operations

ai_index_orgs_users_documents_settings_create

Request

Mixin that includes the StudentTokenAuthentication and IsPlatformAdmin

Security
PlatformApiKeyAuthentication
Path
document_idstringrequired
orgstringrequired
user_idstringrequired
Body
retrain_interval_daysinteger or null[ -2147483648 .. 2147483647 ]
curl -i -X POST \
  'https://docs.ibl.ai/_mock/apis/ibl/api/ai-index/orgs/{org}/users/{user_id}/documents/{document_id}/settings/' \
  -H 'Authorization: YOUR_API_KEY_HERE' \
  -H 'Content-Type: application/json' \
  -d '{
    "retrain_interval_days": -2147483648
  }'

Responses

Bodyapplication/json
retrain_interval_daysinteger or null[ -2147483648 .. 2147483647 ]
Response
application/json
{
  "retrain_interval_days": -2147483648
}

ai_index_orgs_users_documents_graph_train_create

Request

Train a graph-based document through a worker process.

This endpoint queues graph-based documents for training through a specialized worker process that handles knowledge graph processing.

Args: request: The HTTP request containing the document information. org: Organization key identifier.

Returns: Response: A confirmation that the document was queued for training, including a task ID for tracking the progress.

Raises: ValidationError: If the request data is invalid. BadRequest: If there was an error processing the document.

Security
PlatformApiKeyAuthentication
Path
orgstringrequired
user_idstringrequired
Body
required
pathwaystringrequired

Pathway for document to be trained in

urlstring

Url of the document to be trained

textstring

Search text for wikipedia

typestringrequired

Type of document e.g file

translateboolean

If file should be translated

Default false
filestring(binary)

File to be trained

accessstring

Accessibilityto the file

Default "private"
branchstring

Branch of the repository

google_drive_auth_dataany

Authentication and scoped details of google drive

dropbox_auth_dataany

Authentication and scoped details of dropbox

crawler_max_depthinteger

The max depth of the crawler

crawler_max_pages_limitinteger

The max pages limit of the crawler

crawler_max_concurrencyinteger

The max concurrency of the crawler

crawler_match_patternsArray of strings

The patterns that the crawler should use to match urls. Patterns may be a glob pattern or a full regex pattern. Indicate the specified type in crawler_use_glob_matching.

crawler_use_glob_matchingboolean

Whether to use glob patterns for the crawler

curl -i -X POST \
  'https://docs.ibl.ai/_mock/apis/ibl/api/ai-index/orgs/{org}/users/{user_id}/documents/graph-train/' \
  -H 'Authorization: YOUR_API_KEY_HERE' \
  -H 'Content-Type: multipart/form-data' \
  -F pathway=string \
  -F url=string \
  -F text=string \
  -F type=string \
  -F translate=false \
  -F file=string \
  -F access=private \
  -F branch=string \
  -F google_drive_auth_data=null \
  -F dropbox_auth_data=null \
  -F crawler_max_depth=0 \
  -F crawler_max_pages_limit=0 \
  -F crawler_max_concurrency=0 \
  -F crawler_match_patterns=string \
  -F crawler_use_glob_matching=true

Responses

Bodyapplication/json
document_idstring

Document id for document in training

task_idstring

Task id for document in training

messagestringrequired

Message of the training

errorstring
Response
application/json
{
  "task_id": "4194d20c-37d5-4148-882f-f7d2d91f7769",
  "document_id": "doc-123456",
  "message": "Document received. Your request to train has been queued"
}

ai_index_orgs_users_documents_pathways_list

Request

Description: Retrieves a list of document embeddings for a specific pathway with optional search and pagination.

Methods:

  • GET: Retrieves document embeddings that match the specified search criteria and are paginated according to offset and limit.

Parameters:

  • search (str): Search query to filter document names or URLs.
  • offset (int): Offset number for pagination.
  • limit (int): Limit number for pagination.

Returns:

  • GET: A paginated list of document embeddings with their details. { "count": 10, "next": "http://api.example.com/retriever_documents/?offset=10&limit=2", "previous": "http://api.example.com/retriever_documents/?offset=0&limit=2", "results": [ { "document_name": "Document2", "platform_key": "example_platform", "pathway": "example_pathway" } ] }

Error Responses:

  • 400 Bad Request: Invalid query parameters.
  • 404 Not Found: No document embeddings found for the specified criteria.
Security
PlatformApiKeyAuthentication
Path
orgstringrequired
pathwaystringrequired
user_idstringrequired
Query
limitinteger

limit number

offsetinteger

Offset number

searchstringnon-empty

Search query

curl -i -X GET \
  'https://docs.ibl.ai/_mock/apis/ibl/api/ai-index/orgs/{org}/users/{user_id}/documents/pathways/{pathway}/?limit=0&offset=0&search=string' \
  -H 'Authorization: YOUR_API_KEY_HERE'

Responses

Bodyapplication/json
countintegerrequired
Example: 123
nextstring or null(uri)
Example: "http://api.example.org/accounts/?offset=400&limit=100"
previousstring or null(uri)
Example: "http://api.example.org/accounts/?offset=200&limit=100"
resultsArray of objects(RetrieverDocumentEmbedding)required
results[].​idintegerread-onlyrequired
results[].​metadataany or null
results[].​document_namestring or null
results[].​document_typestring or null
results[].​training_status(TrainingStatusEnum (string or null)) or (BlankEnum (any or null)) or (NullEnum (any or null))
One of:
  • trained - Trained
  • failed - Failed
  • pending - Pending
  • untrained - Untrained
string or null(TrainingStatusEnum)
  • trained - Trained
  • failed - Failed
  • pending - Pending
  • untrained - Untrained
Enum"trained""failed""pending""untrained"
results[].​pathwaystringrequired
results[].​urlstring or null
results[].​tokensinteger or null(int64)[ -9223372036854776000 .. 9223372036854776000 ]
results[].​platform_keystringrequired
results[].​is_trainedboolean
results[].​access(AccessEnum (string or null)) or (BlankEnum (any or null)) or (NullEnum (any or null))
One of:
  • public - Public
  • private - Private
string or null(AccessEnum)
  • public - Public
  • private - Private
Enum"public""private"
results[].​crawler_max_depthinteger or null[ -2147483648 .. 2147483647 ]
results[].​crawler_max_pages_limitinteger or null[ -2147483648 .. 2147483647 ]
results[].​crawler_max_concurrencyinteger or null[ -2147483648 .. 2147483647 ]
results[].​crawler_match_patternsany

List of patterns that the crawler should use to match urls. Patterns may be a glob pattern or a full regex pattern. Indicate the specified type in crawler_use_glob_matching.

results[].​crawler_use_glob_matchingboolean
Response
application/json
{
  "count": 123,
  "next": "http://api.example.org/accounts/?offset=400&limit=100",
  "previous": "http://api.example.org/accounts/?offset=200&limit=100",
  "results": [
    [ … ]
  ]
}

ai-marketing

Operations

ai-media

Operations

ai-mentor

Operations

ai-prompt

Operations

career

Operations

catalog

Operations

core

Operations

credentials

Operations

features

Operations

media

Operations

notifications

Operations

scim

Operations

commerce

Operations

recommendations

Operations

reports

Operations

skills

Operations
Copyright © ibl.ai | support@iblai.zendesk.com