Skip to content

Commit d5e090a

Browse files
jexpjexp
committed
Added graphrag-python page
1 parent f1c7b8d commit d5e090a

File tree

12 files changed

+258
-52
lines changed

12 files changed

+258
-52
lines changed

‎modules/genai-ecosystem/nav.adoc‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -7,6 +7,7 @@
77
**** xref:neoconverse.adoc[NeoConverse]
88
**** xref:genai-stack.adoc[GenAI Stack]
99
*** Neo4j GenAI Features
10+
**** xref:graphrag-python.adoc[GraphRAG Python Package]
1011
**** xref:vector-search.adoc[Vector Index and Search]
1112
**** xref:apoc-genai.adoc[APOC GenAI]
1213
*** Cloud Examples

‎modules/genai-ecosystem/pages/_developer_survey.adoc‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
// Developer survey
2+
:page-ad-icon: ~
3+
:page-ad-title: Neo4j Developer Survey
4+
:page-ad-description: Your input matters! Share your Feedback
5+
:page-ad-underline-role: button
6+
:page-ad-underline: Start Here
7+
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey

‎modules/genai-ecosystem/pages/genai-stack.adoc‎

Lines changed: 1 addition & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -7,13 +7,7 @@ include::_graphacademy_llm.adoc[]
77
:neo4j-versions: 5.x
88
:page-pagination:
99
:page-product: genai-stack
10-
// Developer survey
11-
:page-ad-icon: ~
12-
:page-ad-title: Neo4j Developer Survey
13-
:page-ad-description: Your input matters! Share your Feedback
14-
:page-ad-underline-role: button
15-
:page-ad-underline: Start Here
16-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
10+
include::_developer_survey.adoc[]
1711

1812

1913
image::https://dist.neo4j.com/wp-content/uploads/20231005063102/import-embed-data-stack-overflow.png[width=800]

‎modules/genai-ecosystem/pages/graphrag-python.adoc‎

Lines changed: 239 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,239 @@
1+
= Neo4j GraphRAG Python Package
2+
include::_graphacademy_llm.adoc[]
3+
:slug: graphrag-python
4+
:author:
5+
:category: genai-ecosystem
6+
:tags: graphrag, knowledgegraph, embedding, vectorsearch, neo4j, python
7+
:neo4j-versions: 5.23+
8+
:page-pagination:
9+
:page-product: neo4j
10+
11+
12+
The Neo4j GraphRAG package is a comprehensive Python library that allows building GenAI applications.
13+
It supports knowledge graph creation through a pipeline that extracts entities from unstructured text, generates embeddings, and creates a graph in Neo4j.
14+
The package also provides a number of retrievers, for graph search, vector search and integration with vector databases.
15+
16+
== Functionality Includes
17+
18+
* Knowledge Graph Construction Pipeline
19+
* Neo4j Vector Retriever
20+
* Vector Cypher Retriever
21+
* Vector Database Retriever
22+
23+
image::https://cdn.graphacademy.neo4j.com/assets/img/courses/banners/genai-workshop-graphrag.png[width=800,link="https://graphacademy.neo4j.com/courses/genai-workshop-graphrag/"]
24+
25+
== Usage - Examples for a BioMedical Knowledge Graph
26+
27+
First Knowlege Graph Construction using the SimpleKGPipeline
28+
29+
image::https://dist.neo4j.com/wp-content/uploads/20241015075828/simplekgpipeline-1.png[]
30+
31+
Setup of Neo4j connection, schema and foundation models (LLM, Eebeddings) and extraction prompt template.
32+
33+
[source,python]
34+
----
35+
# Neo4j Driver
36+
import neo4j
37+
38+
neo4j_driver = neo4j.GraphDatabase.driver(NEO4J_URI,
39+
auth=(NEO4J_USERNAME, NEO4J_PASSWORD))
40+
41+
# LLM and Embedding Model
42+
from neo4j_graphrag.llm import OpenAILLM
43+
from neo4j_graphrag.embeddings.openai import OpenAIEmbeddings
44+
45+
llm=OpenAILLM(
46+
model_name="gpt-4o-mini",
47+
model_params={
48+
"response_format": {"type": "json_object"}, # use json_object formatting for best results
49+
"temperature": 0 # turning temperature down for more deterministic results
50+
}
51+
)
52+
53+
# Graph Schema Setup
54+
basic_node_labels = ["Object", "Entity", "Group", "Person", "Organization", "Place"]
55+
56+
academic_node_labels = ["ArticleOrPaper", "PublicationOrJournal"]
57+
58+
medical_node_labels = ["Anatomy", "BiologicalProcess", "Cell", "CellularComponent",
59+
"CellType", "Condition", "Disease", "Drug",
60+
"EffectOrPhenotype", "Exposure", "GeneOrProtein", "Molecule",
61+
"MolecularFunction", "Pathway"]
62+
63+
node_labels = basic_node_labels + academic_node_labels + medical_node_labels
64+
65+
# define relationship types
66+
rel_types = ["ACTIVATES", "AFFECTS", "ASSESSES", "ASSOCIATED_WITH", "AUTHORED",
67+
"BIOMARKER_FOR", …]
68+
69+
#create text embedder
70+
embedder = OpenAIEmbeddings()
71+
72+
# define prompt template
73+
prompt_template = '''
74+
You are a medical researcher tasks with extracting information from papers
75+
and structuring it in a property graph to inform further medical and research Q&A.
76+
77+
Extract the entities (nodes) and specify their type from the following Input text.
78+
Also extract the relationships between these nodes. the relationship direction goes from the start node to the end node.
79+
80+
81+
Return result as JSON using the following format:
82+
{{"nodes": [ {{"id": "0", "label": "the type of entity", "properties": {{"name": "name of entity" }} }}],
83+
"relationships": [{{"type": "TYPE_OF_RELATIONSHIP", "start_node_id": "0", "end_node_id": "1", "properties": {{"details": "Description of the relationship"}} }}] }}
84+
85+
...
86+
87+
Use only fhe following nodes and relationships:
88+
{schema}
89+
90+
Assign a unique ID (string) to each node, and reuse it to define relationships.
91+
Do respect the source and target node types for relationship and the relationship direction.
92+
93+
Do not return any additional information other than the JSON in it.
94+
95+
Examples:
96+
{examples}
97+
98+
Input text:
99+
100+
{text}
101+
'''
102+
----
103+
104+
Knowledge Graph Pipeline Setup and Execution with example PDFs
105+
106+
[source,python]
107+
----
108+
# Knowledge Graph Builder
109+
from neo4j_graphrag.experimental.components.text_splitters.fixed_size_splitter import FixedSizeSplitter
110+
from neo4j_graphrag.experimental.pipeline.kg_builder import SimpleKGPipeline
111+
112+
kg_builder_pdf = SimpleKGPipeline(
113+
llm=ex_llm,
114+
driver=driver,
115+
text_splitter=FixedSizeSplitter(chunk_size=500, chunk_overlap=100),
116+
embedder=embedder,
117+
entities=node_labels,
118+
relations=rel_types,
119+
prompt_template=prompt_template,
120+
from_pdf=True
121+
)
122+
123+
pdf_file_paths = ['truncated-pdfs/biomolecules-11-00928-v2-trunc.pdf',
124+
'truncated-pdfs/GAP-between-patients-and-clinicians_2023_Best-Practice-trunc.pdf',
125+
'truncated-pdfs/pgpm-13-39-trunc.pdf']
126+
127+
for path in pdf_file_paths:
128+
print(f"Processing : {path}")
129+
pdf_result = await kg_builder_pdf.run_async(file_path=path)
130+
print(f"Result: {pdf_result}")
131+
----
132+
133+
image::https://dist.neo4j.com/wp-content/uploads/20241015075652/document-chunk-entity.png[width=800]
134+
135+
Then running the GraphRAG Search with the VectorCypher Retriever.
136+
137+
[source,python]
138+
----
139+
from neo4j_graphrag.indexes import create_vector_index
140+
141+
create_vector_index(driver, name="text_embeddings", label="Chunk",
142+
embedding_property="embedding", dimensions=1536, similarity_fn="cosine")
143+
144+
# Vector Retriever
145+
from neo4j_graphrag.retrievers import VectorRetriever
146+
147+
vector_retriever = VectorRetriever(
148+
driver,
149+
index_name="text_embeddings",
150+
embedder=embedder,
151+
return_properties=["text"],
152+
)
153+
154+
# GraphRAG Vector Cypher Retriever
155+
from neo4j_graphrag.retrievers import VectorCypherRetriever
156+
157+
graph_retriever = VectorCypherRetriever(
158+
driver,
159+
index_name="text_embeddings",
160+
embedder=embedder,
161+
retrieval_query="""
162+
//1) Go out 2-3 hops in the entity graph and get relationships
163+
WITH node AS chunk
164+
MATCH (chunk)<-[:FROM_CHUNK]-(entity)-[relList:!FROM_CHUNK]-{1,2}(nb)
165+
UNWIND relList AS rel
166+
167+
//2) collect relationships and text chunks
168+
WITH collect(DISTINCT chunk) AS chunks, collect(DISTINCT rel) AS rels
169+
170+
//3) format and return context
171+
RETURN apoc.text.join([c in chunks | c.text], '\n') +
172+
apoc.text.join([r in rels |
173+
startNode(r).name+' - '+type(r)+' '+r.details+' -> '+endNode(r).name],
174+
'\n') AS info
175+
"""
176+
)
177+
178+
llm = LLM(model_name="gpt-4o", model_params={"temperature": 0.0})
179+
180+
rag_template = RagTemplate(template='''Answer the Question using the following Context. Only respond with information mentioned in the Context. Do not inject any speculative information not mentioned.
181+
182+
# Question:
183+
{query_text}
184+
185+
# Context:
186+
{context}
187+
188+
# Answer:
189+
''', expected_inputs=['query_text', 'context'])
190+
191+
vector_rag = GraphRAG(llm=llm, retriever=vector_retriever, prompt_template=rag_template)
192+
193+
graph_rag = GraphRAG(llm=llm, retriever=graph_retriever, prompt_template=rag_template)
194+
195+
q = "Can you summarize systemic lupus erythematosus (SLE)? including common effects, biomarkers, and treatments? Provide in detailed list format."
196+
197+
vector_rag.search(q, retriever_config={'top_k':5}).answer
198+
graph_rag.search(q, retriever_config={'top_k':5}).answer
199+
----
200+
201+
image::https://dist.neo4j.com/wp-content/uploads/20241128072906/Bildschirmfoto-2024-11-19-um-17.31.45.png[]
202+
203+
== Documentation
204+
205+
[cols="1,4"]
206+
|===
207+
| icon:book[] Documentation | https://neo4j.com/docs/neo4j-graphrag-python/current/
208+
| icon:book[] Guides | https://neo4j.com/docs/neo4j-graphrag-python/current/user_guide_rag.html[RAG & GraphRAG^]
209+
| icon:book[] Guides | https://neo4j.com/docs/neo4j-graphrag-python/current/user_guide_kg_builder.html[Guide Knowledge Graph Builder^]
210+
211+
|===
212+
213+
== Relevant Links
214+
215+
[cols="1,4"]
216+
|===
217+
| icon:user[] Authors | Neo4j Engineering
218+
| icon:github[] Repository | https://github.com/neo4j/neo4j-graphrag-python[GitHub]
219+
| icon:github[] Issues | https://github.com/neo4j/neo4j-graphrag-python/issues
220+
|===
221+
222+
223+
== Videos & Tutorials
224+
225+
++++
226+
<iframe width="560" height="315" src="https://www.youtube.com/embed/hDJlruy60AM?si=TEFW1mj91qrQnaeX" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
227+
++++
228+
229+
++++
230+
<iframe width="560" height="315" src="https://www.youtube.com/embed/OALrsghrP_I?si=Yw08z6fiCp3y_L0j" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
231+
++++
232+
233+
== Highlighted Articles
234+
235+
* https://neo4j.com/blog/graphrag-python-package/[GraphRAG Python Package: Accelerating GenAI With Knowledge Graphs^]
236+
* https://neo4j.com/developer-blog/get-started-graphrag-python-package/[Getting Started With the Neo4j GraphRAG Python Package^]
237+
* https://neo4j.com/developer-blog/graph-traversal-graphrag-python-package/[Vector Search With Graph Traversal the Using Neo4j GraphRAG Package^]
238+
* https://neo4j.com/developer-blog/hybrid-retrieval-graphrag-python-package/[Hybrid Retrieval Using the Neo4j GraphRAG Package for Python^]
239+
* https://neo4j.com/developer-blog/enhancing-hybrid-retrieval-graphrag-python-package/[Enhancing Hybrid Retrieval With Graph Traversal: Neo4j GraphRAG Python^]

‎modules/genai-ecosystem/pages/index.adoc‎

Lines changed: 2 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -8,13 +8,7 @@ include::_graphacademy_llm.adoc[]
88
:neo4j-versions: 5.X
99
:page-pagination:
1010
:page-product: GenAI Ecosystem
11-
// Developer survey
12-
:page-ad-icon: ~
13-
:page-ad-title: Neo4j Developer Survey
14-
:page-ad-description: Your input matters! Share your Feedback
15-
:page-ad-underline-role: button
16-
:page-ad-underline: Start Here
17-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
11+
include::_developer_survey.adoc[]
1812

1913

2014
image::https://dist.neo4j.com/wp-content/uploads/20231030151119/genai-art-diagram-1.svg[width=800]
@@ -79,6 +73,7 @@ Neo4j has worked with the main cloud providers to create GenAI integrations and
7973

8074
Neo4j added a number of features to make it easier to build GenAI applications and integrate LLMs with knowledge graphs.
8175

76+
* xref:graphrag-python.adoc[GraphRAG Package (Python)]
8277
* xref:vector-search.adoc[Vector Index & Search]
8378
* xref:apoc-genai.adoc[APOC GenAI Procedures]
8479

‎modules/genai-ecosystem/pages/langchain.adoc‎

Lines changed: 1 addition & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -7,13 +7,7 @@ include::_graphacademy_llm.adoc[]
77
:neo4j-versions: 5.x
88
:page-pagination:
99
:page-product: langchain
10-
// Developer survey
11-
:page-ad-icon: ~
12-
:page-ad-title: Neo4j Developer Survey
13-
:page-ad-description: Your input matters! Share your Feedback
14-
:page-ad-underline-role: button
15-
:page-ad-underline: Start Here
16-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
10+
include::_developer_survey.adoc[]
1711

1812

1913
image::https://dist.neo4j.com/wp-content/uploads/20230615211357/1AH05dvGA_7db_EMySc9AAw.png[width=800]

‎modules/genai-ecosystem/pages/llamaindex.adoc‎

Lines changed: 1 addition & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -6,14 +6,7 @@
66
:neo4j-versions: 5.x
77
:page-pagination:
88
:page-product: llamaindex
9-
// Developer survey
10-
:page-ad-icon: ~
11-
:page-ad-title: Neo4j Developer Survey
12-
:page-ad-description: Your input matters! Share your Feedback
13-
:page-ad-underline-role: button
14-
:page-ad-underline: Start Here
15-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
16-
9+
include::_developer_survey.adoc[]
1710

1811
// image::todo.png[width=800]
1912

‎modules/genai-ecosystem/pages/llm-graph-builder-deployment.adoc‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -171,7 +171,7 @@ uvicorn score:app --reload
171171
| Env Variable Name | Mandatory/Optional | Default Value | Description
172172
| VITE_BACKEND_API_URL | Optional | http://localhost:8000 | URL for backend API
173173
| VITE_SOURCES | Optional | local,youtube,wiki,s3 | List of input sources that will be available
174-
| VITE_BLOOM_URL | Optional | https://workspace-preview.neo4j.io/workspace/explore?connectURL={CONNECT_URL}&search=Show+me+a+graph | URL for Bloom visualization
174+
| VITE_BLOOM_URL | Optional | https://workspace-preview.neo4j.io/workspace/explore?connectURL=\{CONNECT_URL\}&search=Show+me+a+graph | URL for Bloom visualization
175175
| VITE_LLM_MODELS | Mandatory | diffbot,gpt-4o,gpt-4o-mini | Comma separated list of LLM Model names to show in the selector
176176
|===
177177

‎modules/genai-ecosystem/pages/llm-graph-builder.adoc‎

Lines changed: 1 addition & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -9,13 +9,7 @@ include::_graphacademy_llm.adoc[]
99
:page-pagination:
1010
:page-product: llm-graph-builder
1111
:imagesdir: https://dev.assets.neo4j.com/wp-content/uploads/2024/
12-
// Developer survey
13-
:page-ad-icon: ~
14-
:page-ad-title: Neo4j Developer Survey
15-
:page-ad-description: Your input matters! Share your Feedback
16-
:page-ad-underline-role: button
17-
:page-ad-underline: Start Here
18-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
12+
include::_developer_survey.adoc[]
1913

2014

2115
// image::llm-graph-builder.png[width=600, align=center]

‎modules/genai-ecosystem/pages/rag-demo.adoc‎

Lines changed: 2 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -7,13 +7,8 @@ include::_graphacademy_llm.adoc[]
77
:neo4j-versions: 5.x
88
:page-pagination:
99
:page-product: rag-demo
10-
// Developer survey
11-
:page-ad-icon: ~
12-
:page-ad-title: Neo4j Developer Survey
13-
:page-ad-description: Your input matters! Share your Feedback
14-
:page-ad-underline-role: button
15-
:page-ad-underline: Start Here
16-
:page-ad-link: https://neo4j.typeform.com/to/E6yOZ2Py?utm_source=GA&utm_medium=blurb&utm_campaign=survey
10+
11+
include::_developer_survey.adoc[]
1712

1813

1914
image::https://res.cloudinary.com/dk0tizgdn/image/upload/v1707842287/rag-demo-short_vwezew.gif[width=600, align=center]

0 commit comments

Comments
 (0)