Workflows in the Clarifai API

Setting Up Mesh Workflows

Create

To create a new custom workflow, specify a list of model IDs that are to be included in the workflow. Each model ID also requires a specific model version ID, since a model can have several versions.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

post_workflows_response = stub.PostWorkflows(
    service_pb2.PostWorkflowsRequest(
      user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
      workflows=[
        resources_pb2.Workflow(
          id="my-custom-workflow",
          nodes=[
            resources_pb2.WorkflowNode(
              id="food-concepts",
              model=resources_pb2.Model(
                id="bd367be194cf45149e75f01d59f77ba7",
                model_version=resources_pb2.ModelVersion(
                  id="dfebc169854e429086aceb8368662641"
                )
              )
            ),
            resources_pb2.WorkflowNode(
              id="general-concepts",
              model=resources_pb2.Model(
                id="aaa03c23b3724a16a56b629203edc62c",
                model_version=resources_pb2.ModelVersion(
                  id="aa9ca48295b37401f8af92ad1af0d91d"
                )
              )
            ),
          ]
        )
      ]
    ),
    metadata=metadata
)

if post_workflows_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Post workflows failed, status: " + post_workflows_response.status.description)

 

cURL

# The first model is the Clarifai's Food model, and the second the Clarifai's General model.

curl -X POST 'https://api.clarifai.com/v2/workflows' \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY' \
    --data-raw '{
      "workflows": [{
        "id": "my-custom-workflow",
        "nodes": [
          {
            "id": "food-concepts",
            "model": {
              "id": "bd367be194cf45149e75f01d59f77ba7",
              "model_version": {
                "id": "dfebc169854e429086aceb8368662641"
              }
            }
          },
          {
            "id": "general-concepts",
            "model": {
              "id": "aaa03c23b3724a16a56b629203edc62c",
              "model_version": {
                "id": "aa9ca48295b37401f8af92ad1af0d91d"
              }
            }
          }
        ]
      }]
    }'

 

Workflow Predict

Predict using a workflow. The response will contain the predictions each model in the workflow returns for the input.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

post_workflow_results_response = stub.PostWorkflowResults(
    service_pb2.PostWorkflowResultsRequest(
        user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
        workflow_id="{YOUR_WORKFLOW_ID}",
        inputs=[
            resources_pb2.Input(
                data=resources_pb2.Data(
                    image=resources_pb2.Image(
                        url="https://samples.clarifai.com/metro-north.jpg"
                    )
                )
            )
        ]
    ),
    metadata=metadata
)
if post_workflow_results_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Post workflow results failed, status: " + post_workflow_results_response.status.description)

# We'll get one WorkflowResult for each input we used above. Because of one input, we have here
# one WorkflowResult.
results = post_workflow_results_response.results[0]

# Each model we have in the workflow will produce one output.
for output in results.outputs:
    model = output.model

    print("Predicted concepts for the model `%s`" % model.name)
    for concept in output.data.concepts:
        print("\t%s %.2f" % (concept.name, concept.value))

 

cURL

curl -XPOST \
  -H'authorization: KeyYOUR_API_KEY' \
  -H'content-type: application/json' \
  -d '{
    "inputs": [
        {
          "data": {
            "image": {
              "url": "https://samples.clarifai.com/metro-north.jpg"
          }
        }
      }
    ]
}'\
https://api.clarifai.com/v2/workflows/{YOUR_WORKFLOW_ID}/results

 

Get

Get all workflows in an app

Return all custom workflows in your app.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

list_workflows_response = stub.ListWorkflows(
    service_pb2.ListWorkflowsRequest(
        user_app_id=userDataObject  # The userDataObject is created in the overview and is required when using a PAT
    ),
    metadata=metadata
)

if list_workflows_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("List workflows failed, status: " + list_workflows_response.status.description)

for workflow in list_workflows_response.workflows:
    print(f"The workflow {workflow.id} consists of these models:")
    for workflow_node in workflow.nodes:
        model = workflow_node.model
        print(model.id)
    print()

 

cURL

curl -X GET 'https://api.clarifai.com/v2/workflows' \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY'

 

Get a workflow by a specific ID

Returns information about a specific workflow.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

get_workflow_response = stub.GetWorkflow(
    service_pb2.GetWorkflowRequest(
        user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
        workflow_id="my-custom-workflow"
    ),
    metadata=metadata
)

if get_workflow_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Get workflow failed, status: " + get_workflow_response.status.description)

workflow = get_workflow_response.workflow
print(f"The workflow consists of these models:")
for workflow_node in workflow.nodes:
    model = workflow_node.model
    print(model.id)

 

cURL

curl -X GET 'https://api.clarifai.com/v2/workflows/my-custom-workflow' \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY'

 

Update

Patch workflow

Ability to change the workflow, that is to change the models of which the workflow consists.

Possible actions are "overwrite", "merge" and "remove".

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

patch_workflows_response = stub.PatchWorkflows(
    service_pb2.PatchWorkflowsRequest(
      user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
      action="overwrite",
      workflows=[
        resources_pb2.Workflow(
          id="my-custom-workflow",
          nodes=[
            resources_pb2.WorkflowNode(
              id="travel-concepts",
              model=resources_pb2.Model(
                id="ccc28c313d69466f836ab83287a54ed9",
                model_version=resources_pb2.ModelVersion(
                  id="cce28c313d69466f836ab83287a54ed9"
                )
              )
            ),
            resources_pb2.WorkflowNode(
              id="nsfw-concepts",
              model=resources_pb2.Model(
                id="ccc76d86d2004ed1a38ba0cf39ecb4b1",
                model_version=resources_pb2.ModelVersion(
                  id="cc76a92beaeb4d8495a58ba197998158"
                )
              )
            ),
            resources_pb2.WorkflowNode(
              id="wedding-concepts",
              model=resources_pb2.Model(
                id="c386b7a870114f4a87477c0824499348",
                model_version=resources_pb2.ModelVersion(
                  id="787cc9a002164250800598d36b072384"
                )
              )
            ),
          ]
        )
      ]
    ),
    metadata=metadata
)

if patch_workflows_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Patch workflows failed, status: " + patch_workflows_response.status.description)

 

cURL

# Supported actions are: overwrite, merge, remove.

curl -X PATCH 'https://api.clarifai.com/v2/workflows' \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY' \
    --data-raw '{
        "action": "overwrite",
        "workflows": [
            {
                "id": "my-custom-workflow",
                "nodes": [
                    {
                        "id": "travel-concepts",
                        "model": {
                            "id": "ccc28c313d69466f836ab83287a54ed9",
                            "model_version": {
                                "id": "cce28c313d69466f836ab83287a54ed9"
                            }
                        }
                    },
                    {
                        "id": "nsfw-concepts",
                        "model": {
                            "id": "ccc76d86d2004ed1a38ba0cf39ecb4b1",
                            "model_version": {
                                "id": "cc76a92beaeb4d8495a58ba197998158"
                            }
                        }
                    },
                    {
                        "id": "wedding-concepts",
                        "model": {
                            "id": "c386b7a870114f4a87477c0824499348",
                            "model_version": {
                                "id": "787cc9a002164250800598d36b072384"
                            }
                        }
                    }
                ]
            }
        ]
    }'

 

Delete

Delete workflow by ID

Delete a specific workflow.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

delete_workflow_response = stub.DeleteWorkflow(
    service_pb2.DeleteWorkflowRequest(
      user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
      workflow_id="my-custom-workflow"
    ),
    metadata=metadata
)

if delete_workflow_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Delete workflow failed, status: " + delete_workflow_response.status.description)

 

cURL

curl -X DELETE 'https://api.clarifai.com/v2/workflows/my-custom-workflow \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY'

 

Delete all workflows

Deletes all custom workflows.

Note: instead of "delete_all" it's possible to specify a list of workflow IDs to be deleted, using the ids field.

 

Python

# Insert here the initialization code as outlined on this page:
# https://docs.clarifai.com/api-guide/api-overview/api-clients#client-installation-instructions

delete_workflows_response = stub.DeleteWorkflows(
    service_pb2.DeleteWorkflowsRequest(
      user_app_id=userDataObject,  # The userDataObject is created in the overview and is required when using a PAT
      delete_all=True
    ),
    metadata=metadata
)

if delete_workflows_response.status.code != status_code_pb2.SUCCESS:
    raise Exception("Delete workflows failed, status: " + delete_workflows_response.status.description)

 

cURL

curl -X DELETE 'https://api.clarifai.com/v2/workflows' \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Key YOUR_API_KEY' \
    --data-raw '{
        "delete_all": true
    }'