Provenance integration with 3rd party libraries

Provena facilitates the capture of provenance primarily using a specialisation of Prov-O. We are able to interface with Provena provenance information using graph based tools. This guide provides some ideas for fetching provenance information from the Provena Registry and using prov and rdflib python libraries.

This guide will show how you can use prov and rdflib to serialise into PROV documents and RDF formats. This is useful if you would like to use those toolsets to query, integrate with other PROV/RDF data, or visualise using tools such as GraphViz.

Pre-requisites will be:

rdflib
prov
graphviz

Using Prov and rdflib python libraries

First initialise the provena client.

from provenaclient import ProvenaClient, Config
from provenaclient.auth import DeviceFlow
from provenaclient.auth.implementations import OfflineFlow
from ProvenaInterfaces.RegistryAPI import *
from ProvenaInterfaces.RegistryModels import *
client_config = Config(
        domain="dev.rrap-is.com",
        realm_name="rrap"
    )
auth = DeviceFlow(config=client_config, client_id="client-tools")

client = ProvenaClient(auth=auth, config=client_config)

Query all model run records. Prov compatible data is currently available from Provena primarily from model run records as it specialises from prov:Activity and relates relevant Entities and Agents via that construct.

# Get model run record in the Provena registry
model_run_record_list = await client.registry.model_run.list_items(GeneralListRequest(
    filter_by=None,
    sort_by=None,
    pagination_key=None
))

model_run_record = model_run_record_list.items.pop()

We can instantiate the PROV compatible serialisation of the model run into a ProvDocument using the Prov python library. This will allow us to interact natively with the provenance document in Python.

import prov.model as pm

# Get a Prov document from the record
document = pm.ProvDocument.deserialize(source=None, content=model_run_record.prov_serialisation)

The prov representation of the model run record is quite minimal. The available attributes are: id, item_category, item_subtype. However, from this information alone, we can enrich the prov document/graph with additional info by fetching it from the Registry.

We show the use of general_fetch_item(), which returns general info such as the label. However, you can use the more specialised fetch() in the provena client that allows more specific info to be fetched from the item, e.g. Model details or Dataset metadata.

# Resolve labels by fetching items from the registry
for r in document.get_records():   
    if r.get_attribute('id'): #only fetch labels if the record has an ID attribute
        id = next(iter(r.get_attribute('id')))
        item_category = next(iter(r.get_attribute('item_category')))
        item_subtype = next(iter(r.get_attribute('item_subtype')))
        print(f"{id} {r.get_type()} {item_category} {item_subtype}")
        res = await client.registry.general_fetch_item(id=id)
        print(f"{res.item['display_name']} {res.item['item_subtype']}")
        
        #Add the label to the prov doc for this record
        r.add_attributes( {'prov:label': res.item['display_name']} )

10378.1/1964491 prov:Activity ACTIVITY MODEL_RUN
study example MODEL_RUN
10378.1/1964490 prov:Activity ACTIVITY STUDY
study 2 STUDY
10378.1/1963761 prov:Entity ENTITY WORKFLOW_TEMPLATE
test MODEL_RUN_WORKFLOW_TEMPLATE
10378.1/1963759 prov:Entity ENTITY MODEL
test MODEL
10378.1/1963764 prov:Entity ENTITY DATASET
output DATASET
10378.1/1963760 prov:Entity ENTITY DATASET_TEMPLATE
input DATASET_TEMPLATE
10378.1/1963721 prov:Entity ENTITY DATASET
FIRST dataset DATASET
10378.1/1963718 prov:Agent AGENT PERSON
Peter 0.1x-dev Baker PERSON
10378.1/1963720 prov:Agent AGENT ORGANISATION
CSIRO ORGANISATION

# Serialise this to RDF/Turtle using underlying rdflib 
print(document.serialize(format='rdf', rdf_format='ttl'))

@prefix ns1: <http://hdl.handle.net/> .
@prefix prov: <http://www.w3.org/ns/prov#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .

<http://hdl.handle.net/10378.1/1963764> a prov:Entity ;
    rdfs:label "output"^^xsd:string ;
    ns1:id "10378.1/1963764"^^xsd:string ;
    ns1:item_category "ENTITY"^^xsd:string ;
    ns1:item_subtype "DATASET"^^xsd:string ;
    prov:wasAttributedTo <http://hdl.handle.net/10378.1/1963718> ;
    prov:wasGeneratedBy <http://hdl.handle.net/10378.1/1964491> ;
    prov:wasInfluencedBy <http://hdl.handle.net/10378.1/1963760> .

<http://hdl.handle.net/10378.1/1963720> a prov:Agent ;
    rdfs:label "CSIRO"^^xsd:string ;
    ns1:id "10378.1/1963720"^^xsd:string ;
    ns1:item_category "AGENT"^^xsd:string ;
    ns1:item_subtype "ORGANISATION"^^xsd:string .

<http://hdl.handle.net/10378.1/1963721> a prov:Entity ;
    rdfs:label "FIRST dataset"^^xsd:string ;
    ns1:id "10378.1/1963721"^^xsd:string ;
    ns1:item_category "ENTITY"^^xsd:string ;
    ns1:item_subtype "DATASET"^^xsd:string ;
    prov:wasInfluencedBy <http://hdl.handle.net/10378.1/1963760> .

<http://hdl.handle.net/10378.1/1963761> a prov:Entity ;
    rdfs:label "test"^^xsd:string ;
    ns1:id "10378.1/1963761"^^xsd:string ;
    ns1:item_category "ENTITY"^^xsd:string ;
    ns1:item_subtype "WORKFLOW_TEMPLATE"^^xsd:string ;
    prov:hadMember <http://hdl.handle.net/10378.1/1963759>,
        <http://hdl.handle.net/10378.1/1963760> .

<http://hdl.handle.net/10378.1/1964490> a prov:Activity ;
    rdfs:label "study 2"^^xsd:string ;
    ns1:id "10378.1/1964490"^^xsd:string ;
    ns1:item_category "ACTIVITY"^^xsd:string ;
    ns1:item_subtype "STUDY"^^xsd:string .

<http://hdl.handle.net/10378.1/1964491> a prov:Activity ;
    rdfs:label "study example"^^xsd:string ;
    ns1:id "10378.1/1964491"^^xsd:string ;
    ns1:item_category "ACTIVITY"^^xsd:string ;
    ns1:item_subtype "MODEL_RUN"^^xsd:string ;
    prov:used <http://hdl.handle.net/10378.1/1963721>,
        <http://hdl.handle.net/10378.1/1963759>,
        <http://hdl.handle.net/10378.1/1963761> ;
    prov:wasAssociatedWith <http://hdl.handle.net/10378.1/1963718>,
        <http://hdl.handle.net/10378.1/1963720> ;
    prov:wasInformedBy <http://hdl.handle.net/10378.1/1964490> .

<http://hdl.handle.net/10378.1/1963718> a prov:Agent ;
    rdfs:label "Peter 0.1x-dev Baker"^^xsd:string ;
    ns1:id "10378.1/1963718"^^xsd:string ;
    ns1:item_category "AGENT"^^xsd:string ;
    ns1:item_subtype "PERSON"^^xsd:string .

<http://hdl.handle.net/10378.1/1963759> a prov:Entity ;
    rdfs:label "test"^^xsd:string ;
    ns1:id "10378.1/1963759"^^xsd:string ;
    ns1:item_category "ENTITY"^^xsd:string ;
    ns1:item_subtype "MODEL"^^xsd:string .

<http://hdl.handle.net/10378.1/1963760> a prov:Entity ;
    rdfs:label "input"^^xsd:string ;
    ns1:id "10378.1/1963760"^^xsd:string ;
    ns1:item_category "ENTITY"^^xsd:string ;
    ns1:item_subtype "DATASET_TEMPLATE"^^xsd:string .

# other prov syntax
print(document.get_provn())

document
  default <http://hdl.handle.net/>
  
  activity(10378.1/1964491, -, -, [id="10378.1/1964491", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="study example"])
  activity(10378.1/1964490, -, -, [id="10378.1/1964490", item_category="ACTIVITY", item_subtype="STUDY", prov:label="study 2"])
  entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
  entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
  entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
  entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
  entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
  agent(10378.1/1963718, [id="10378.1/1963718", item_category="AGENT", item_subtype="PERSON", prov:label="Peter 0.1x-dev Baker"])
  agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
  wasInformedBy(10378.1/1964491, 10378.1/1964490)
  used(10378.1/1964491, 10378.1/1963721, -)
  used(10378.1/1964491, 10378.1/1963759, -)
  used(10378.1/1964491, 10378.1/1963761, -)
  wasGeneratedBy(10378.1/1963764, 10378.1/1964491, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963718, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963720, -)
  wasAttributedTo(10378.1/1963764, 10378.1/1963718)
  hadMember(10378.1/1963761, 10378.1/1963760)
  hadMember(10378.1/1963761, 10378.1/1963759)
  wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
  wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
endDocument

Visualisation using Graphviz

from prov.dot import prov_to_dot
from IPython.display import Image

# render into file
dot = prov_to_dot(document)
name = "prov_graph"

dot.write_png(name + '.png')
Image(name + '.png')

_images/8bde08526b47e49113fa905b46ae8528a45168090f6e549b4c82c9e7e16f5a89.png

Advanced use case: Querying and joining graphs

You may want to build out a graph from your starting point. The following shows how we can expand the provenance graph from a starting point.

from provenaclient.models.general import CustomLineageResponse
from prov.model import ProvDocument
#let's use the previous model run graph as a starting point

#fetch upstream nodes from 
upstream_result = await client.prov_api.explore_upstream(model_run_record.id)

for node in upstream_result.graph.nodes:
    if node.item_subtype == ItemSubType.MODEL_RUN:
        print(f"{node}")
        #fetch and add prov doc to existing doc
        mr = await client.registry.model_run.fetch(id=node.id)
        doc = pm.ProvDocument.deserialize(source=None, content=mr.item.prov_serialisation)
        records = doc.get_records()
        for r in records:
            document.add_record(r)
            if r.get_attribute('id'):
                id = next(iter(r.get_attribute('id')))
                res = await client.registry.general_fetch_item(id=id)
                r.add_attributes( {'prov:label': res.item['display_name']} )
            print(r)

        
print(document.get_provn())

id='10378.1/1964491' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
activity(10378.1/1964491, -, -, [id="10378.1/1964491", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="study example"])
activity(10378.1/1964490, -, -, [id="10378.1/1964490", item_category="ACTIVITY", item_subtype="STUDY", prov:label="study 2"])
entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
agent(10378.1/1963718, [id="10378.1/1963718", item_category="AGENT", item_subtype="PERSON", prov:label="Peter 0.1x-dev Baker"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
wasInformedBy(10378.1/1964491, 10378.1/1964490)
used(10378.1/1964491, 10378.1/1963721, -)
used(10378.1/1964491, 10378.1/1963759, -)
used(10378.1/1964491, 10378.1/1963761, -)
wasGeneratedBy(10378.1/1963764, 10378.1/1964491, -)
wasAssociatedWith(10378.1/1964491, 10378.1/1963718, -)
wasAssociatedWith(10378.1/1964491, 10378.1/1963720, -)
wasAttributedTo(10378.1/1963764, 10378.1/1963718)
hadMember(10378.1/1963761, 10378.1/1963760)
hadMember(10378.1/1963761, 10378.1/1963759)
wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
document
  default <http://hdl.handle.net/>
  
  activity(10378.1/1964491, -, -, [id="10378.1/1964491", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="study example"])
  activity(10378.1/1964490, -, -, [id="10378.1/1964490", item_category="ACTIVITY", item_subtype="STUDY", prov:label="study 2"])
  entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
  entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
  entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
  entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
  entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
  agent(10378.1/1963718, [id="10378.1/1963718", item_category="AGENT", item_subtype="PERSON", prov:label="Peter 0.1x-dev Baker"])
  agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
  wasInformedBy(10378.1/1964491, 10378.1/1964490)
  used(10378.1/1964491, 10378.1/1963721, -)
  used(10378.1/1964491, 10378.1/1963759, -)
  used(10378.1/1964491, 10378.1/1963761, -)
  wasGeneratedBy(10378.1/1963764, 10378.1/1964491, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963718, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963720, -)
  wasAttributedTo(10378.1/1963764, 10378.1/1963718)
  hadMember(10378.1/1963761, 10378.1/1963760)
  hadMember(10378.1/1963761, 10378.1/1963759)
  wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
  wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
  activity(10378.1/1964491, -, -, [id="10378.1/1964491", item_category="ACTIVITY", item_subtype="MODEL_RUN"])
  activity(10378.1/1964490, -, -, [id="10378.1/1964490", item_category="ACTIVITY", item_subtype="STUDY"])
  entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE"])
  entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL"])
  entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET"])
  entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE"])
  entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET"])
  agent(10378.1/1963718, [id="10378.1/1963718", item_category="AGENT", item_subtype="PERSON"])
  agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION"])
  wasInformedBy(10378.1/1964491, 10378.1/1964490)
  used(10378.1/1964491, 10378.1/1963721, -)
  used(10378.1/1964491, 10378.1/1963759, -)
  used(10378.1/1964491, 10378.1/1963761, -)
  wasGeneratedBy(10378.1/1963764, 10378.1/1964491, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963718, -)
  wasAssociatedWith(10378.1/1964491, 10378.1/1963720, -)
  wasAttributedTo(10378.1/1963764, 10378.1/1963718)
  hadMember(10378.1/1963761, 10378.1/1963760)
  hadMember(10378.1/1963761, 10378.1/1963759)
  wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
  wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
endDocument

Visualise the prov graph using GraphViz

from prov.dot import prov_to_dot
from IPython.display import Image

# render into file
dot = prov_to_dot(document)
name = "prov_graph2"

dot.write_png(name + '.png')
Image(name + '.png')

_images/66c0765c8c19eebcd070ef612d6769fd63bb9f1e94b441cc712abbb403867055.png

Let’s look at one of the input datasets and get all upstream and downstream things from it

dataset_id = "10378.1/1963721" #this will be different for your context

#fetch upstream nodes from 
upstream_result = await client.prov_api.explore_upstream(dataset_id)
downstream_result = await client.prov_api.explore_downstream(dataset_id)

for node in upstream_result.graph.nodes:
    if node.item_subtype == ItemSubType.MODEL_RUN:
        print(f"{node}")
        #fetch and add prov doc to existing doc
        mr = await client.registry.model_run.fetch(id=node.id)
        doc = pm.ProvDocument.deserialize(source=None, content=mr.item.prov_serialisation)
        records = doc.get_records()
        for r in records:
            document.add_record(r)
            if r.get_attribute('id'):
                id = next(iter(r.get_attribute('id')))
                res = await client.registry.general_fetch_item(id=id)
                r.add_attributes( {'prov:label': res.item['display_name']} )
            print(r)
for node in downstream_result.graph.nodes:
    if node.item_subtype == ItemSubType.MODEL_RUN:
        print(f"{node}")
        #fetch and add prov doc to existing doc
        mr = await client.registry.model_run.fetch(id=node.id)
        doc = pm.ProvDocument.deserialize(source=None, content=mr.item.prov_serialisation)
        records = doc.get_records()
        for r in records:
            document.add_record(r)
            if r.get_attribute('id'):
                id = next(iter(r.get_attribute('id')))
                res = await client.registry.general_fetch_item(id=id)
                r.add_attributes( {'prov:label': res.item['display_name']} )
            print(r)

id='10378.1/1964491' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
activity(10378.1/1964491, -, -, [id="10378.1/1964491", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="study example"])
activity(10378.1/1964490, -, -, [id="10378.1/1964490", item_category="ACTIVITY", item_subtype="STUDY", prov:label="study 2"])
entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
agent(10378.1/1963718, [id="10378.1/1963718", item_category="AGENT", item_subtype="PERSON", prov:label="Peter 0.1x-dev Baker"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
wasInformedBy(10378.1/1964491, 10378.1/1964490)
used(10378.1/1964491, 10378.1/1963721, -)
used(10378.1/1964491, 10378.1/1963759, -)
used(10378.1/1964491, 10378.1/1963761, -)
wasGeneratedBy(10378.1/1963764, 10378.1/1964491, -)
wasAssociatedWith(10378.1/1964491, 10378.1/1963718, -)
wasAssociatedWith(10378.1/1964491, 10378.1/1963720, -)
wasAttributedTo(10378.1/1963764, 10378.1/1963718)
hadMember(10378.1/1963761, 10378.1/1963760)
hadMember(10378.1/1963761, 10378.1/1963759)
wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
id='10378.1/1964549' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
agent(10378.1/1964548, [id="10378.1/1964548", item_category="AGENT", item_subtype="PERSON", prov:label="The Breaking Person"])
activity(10378.1/1964549, -, -, [id="10378.1/1964549", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="Test Model Run to see if this works 😊"])
used(10378.1/1964549, 10378.1/1963721, -)
used(10378.1/1964549, 10378.1/1963759, -)
used(10378.1/1964549, 10378.1/1963761, -)
wasGeneratedBy(10378.1/1963764, 10378.1/1964549, -)
wasAssociatedWith(10378.1/1964549, 10378.1/1964548, -)
wasAssociatedWith(10378.1/1964549, 10378.1/1963720, -)
wasAttributedTo(10378.1/1963764, 10378.1/1964548)
hadMember(10378.1/1963761, 10378.1/1963760)
hadMember(10378.1/1963761, 10378.1/1963759)
wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
id='10378.1/1964550' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
activity(10378.1/1963758, -, -, [id="10378.1/1963758", item_category="ACTIVITY", item_subtype="STUDY", prov:label="test"])
activity(10378.1/1964550, -, -, [id="10378.1/1964550", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="Test Model Run to see if this works 😊"])
entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
agent(10378.1/1964548, [id="10378.1/1964548", item_category="AGENT", item_subtype="PERSON", prov:label="The Breaking Person"])
wasInformedBy(10378.1/1964550, 10378.1/1963758)
used(10378.1/1964550, 10378.1/1963721, -)
used(10378.1/1964550, 10378.1/1963759, -)
used(10378.1/1964550, 10378.1/1963761, -)
wasGeneratedBy(10378.1/1963764, 10378.1/1964550, -)
wasAssociatedWith(10378.1/1964550, 10378.1/1964548, -)
wasAssociatedWith(10378.1/1964550, 10378.1/1963720, -)
wasAttributedTo(10378.1/1963764, 10378.1/1964548)
hadMember(10378.1/1963761, 10378.1/1963760)
hadMember(10378.1/1963761, 10378.1/1963759)
wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
wasInfluencedBy(10378.1/1963764, 10378.1/1963760)
id='10378.1/1973180' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
entity(10378.1/1972856, [id="10378.1/1972856", item_category="ENTITY", item_subtype="DATASET", prov:label="Zarr - Test Dataset"])
entity(10378.1/1972891, [id="10378.1/1972891", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="Output dataset template"])
entity(10378.1/1972888, [id="10378.1/1972888", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="Fire model workflow template"])
entity(10378.1/1972876, [id="10378.1/1972876", item_category="ENTITY", item_subtype="MODEL", prov:label="Fire spread model"])
entity(10378.1/1972890, [id="10378.1/1972890", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="Input dataset template"])
activity(10378.1/1973180, -, -, [id="10378.1/1973180", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="Notebook Model Run Testing"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
agent(10378.1/1968384, [id="10378.1/1968384", item_category="AGENT", item_subtype="PERSON", prov:label="Jonathan Yu"])
used(10378.1/1973180, 10378.1/1963721, -)
used(10378.1/1973180, 10378.1/1972876, -)
used(10378.1/1973180, 10378.1/1972888, -)
wasGeneratedBy(10378.1/1972856, 10378.1/1973180, -)
wasAssociatedWith(10378.1/1973180, 10378.1/1968384, -)
wasAssociatedWith(10378.1/1973180, 10378.1/1963720, -)
wasAttributedTo(10378.1/1972856, 10378.1/1968384)
hadMember(10378.1/1972888, 10378.1/1972890)
hadMember(10378.1/1972888, 10378.1/1972891)
hadMember(10378.1/1972888, 10378.1/1972876)
wasInfluencedBy(10378.1/1963721, 10378.1/1972890)
wasInfluencedBy(10378.1/1972856, 10378.1/1972891)
id='10378.1/1963767' item_category=<ItemCategory.ACTIVITY: 'ACTIVITY'> item_subtype=<ItemSubType.MODEL_RUN: 'MODEL_RUN'>
entity(10378.1/1963760, [id="10378.1/1963760", item_category="ENTITY", item_subtype="DATASET_TEMPLATE", prov:label="input"])
entity(10378.1/1963721, [id="10378.1/1963721", item_category="ENTITY", item_subtype="DATASET", prov:label="FIRST dataset"])
entity(10378.1/1963764, [id="10378.1/1963764", item_category="ENTITY", item_subtype="DATASET", prov:label="output"])
entity(10378.1/1963759, [id="10378.1/1963759", item_category="ENTITY", item_subtype="MODEL", prov:label="test"])
entity(10378.1/1963761, [id="10378.1/1963761", item_category="ENTITY", item_subtype="WORKFLOW_TEMPLATE", prov:label="test"])
activity(10378.1/1965693, -, -, [id="10378.1/1965693", item_category="ACTIVITY", item_subtype="STUDY", prov:label="Jordan's Study of All Things Provena"])
activity(10378.1/1963767, -, -, [id="10378.1/1963767", item_category="ACTIVITY", item_subtype="MODEL_RUN", prov:label="test"])
agent(10378.1/1963719, [id="10378.1/1963719", item_category="AGENT", item_subtype="PERSON", prov:label="Ross 10x-dev Petridis"])
agent(10378.1/1963720, [id="10378.1/1963720", item_category="AGENT", item_subtype="ORGANISATION", prov:label="CSIRO"])
wasInformedBy(10378.1/1963767, 10378.1/1965693)
used(10378.1/1963767, 10378.1/1963721, -)
used(10378.1/1963767, 10378.1/1963759, -)
used(10378.1/1963767, 10378.1/1963761, -)
wasGeneratedBy(10378.1/1963764, 10378.1/1963767, -)
wasAssociatedWith(10378.1/1963767, 10378.1/1963719, -)
wasAssociatedWith(10378.1/1963767, 10378.1/1963720, -)
wasAttributedTo(10378.1/1963764, 10378.1/1963719)
hadMember(10378.1/1963761, 10378.1/1963760)
hadMember(10378.1/1963761, 10378.1/1963759)
wasInfluencedBy(10378.1/1963721, 10378.1/1963760)
wasInfluencedBy(10378.1/1963764, 10378.1/1963760)

# render into file
dot = prov_to_dot(document)
name = "prov_graph3"

dot.write_png(name + '.png')
Image(name + '.png')

_images/5c97ac423cfc92424e98390cb30df9db42ae5ef843c8286ed8cea5184c3a3729.png