Knowledge Graph Extraction¶

from llcuda.api import LlamaCppClient
from llcuda.graphistry import SplitGPUManager

manager = SplitGPUManager()
manager.assign_llm(0)

client = LlamaCppClient()
response = client.chat.completions.create(
    messages=[{
        "role": "user",
        "content": "Extract entities and relationships from: ..."
    }]
)

text = response.choices[0].message.content

import json

# Parse LLM output
data = json.loads(text)
entities = data['entities']
relationships = data['relationships']

import cudf

nodes_df = cudf.DataFrame(entities)
edges_df = cudf.DataFrame(relationships)

from llcuda.graphistry import GraphWorkload, register_graphistry

workload = GraphWorkload(gpu_id=1)
register_graphistry(api=3, protocol="https", server="hub.graphistry.com")

g = workload.create_knowledge_graph(
    entities=entities,
    relationships=relationships
)
g.plot()