feat: integration on pinecone hosted rerankers (#17192)

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/.gitignore

@@ -0,0 +1,153 @@
+llama_index/_static
+.DS_Store
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+bin/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+etc/
+include/
+lib/
+lib64/
+parts/
+sdist/
+share/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+.ruff_cache
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+notebooks/
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+pyvenv.cfg
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Jetbrains
+.idea
+modules/
+*.swp
+
+# VsCode
+.vscode
+
+# pipenv
+Pipfile
+Pipfile.lock
+
+# pyright
+pyrightconfig.json

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/BUILD

@@ -0,0 +1,3 @@
+poetry_requirements(
+    name = "poetry",
+)

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/Makefile

@@ -0,0 +1,17 @@
+GIT_ROOT ?= $(shell git rev-parse --show-toplevel)
+
+help:	## Show all Makefile targets.
+	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[33m%-30s\033[0m %s\n", $$1, $$2}'
+
+format:	## Run code autoformatters (black).
+	pre-commit install
+	git ls-files | xargs pre-commit run black --files
+
+lint:	## Run linters: pre-commit (black, ruff, codespell) and mypy
+	pre-commit install && git ls-files | xargs pre-commit run --show-diff-on-failure --files
+
+test:	## Run tests via pytest.
+	pytest tests
+
+watch-docs:	## Build and watch documentation.
+	sphinx-autobuild docs/ docs/_build/html --open-browser --watch $(GIT_ROOT)/llama_index/

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/README.md

@@ -0,0 +1,73 @@
+# LLamaIndex node postprocessing reranker using pinecone hosted models
+
+- use rerank models with the pinecone managed vector service to rerank the search results
+- available rerank models from [pinecone](https://app.pinecone.io/organizations/-Nn577_974iRsvC6nVxg/projects/a4fe57a4-b1cc-4a99-bf1d-c35a595cae4a/models)
+
+```python
+import os
+from llama_index.core import VectorStoreIndex
+from llama_index.core.schema import TextNode
+from llama_index.postprocessor.pinecone_native_rerank import (
+    PineconeNativeRerank,
+)
+from llama_index.core.response.pprint_utils import pprint_response
+
+os.environ["PINECONE_API_KEY"] = "your_pinecone_api_key"
+os.environ["OPENAI_API_KEY"] = "your_openai_api_key"
+
+txts = [
+    "Apple is a popular fruit known for its sweetness and crisp texture.",
+    "Apple is known for its innovative products like the iPhone.",
+    "Many people enjoy eating apples as a healthy snack.",
+    "Apple Inc. has revolutionized the tech industry with its sleek designs and user-friendly interfaces.",
+    "An apple a day keeps the doctor away, as the saying goes.",
+    "apple has a lot of vitamins",
+]
+
+nodes = [TextNode(id_=f"vec{i}", text=txt) for i, txt in enumerate(txts)]
+
+pinecone_reranker = PineconeNativeRerank(top_n=4, model="pinecone-rerank-v0")
+
+index = VectorStoreIndex(nodes)
+
+query_engine = index.as_query_engine(
+    similarity_top_k=10,
+    node_postprocessors=[pinecone_reranker],
+)
+
+response = query_engine.query(
+    "The tech company Apple is known for its innovative products like the iPhone."
+)
+
+pprint_response(response, show_source=True)
+```
+
+output
+
+```txt
+Final Response: Apple is recognized for its innovative products like
+the iPhone.
+______________________________________________________________________
+Source Node 1/4
+Node ID: vec1
+Similarity: 0.9655668
+Text: Apple is known for its innovative products like the iPhone.
+______________________________________________________________________
+Source Node 2/4
+Node ID: vec3
+Similarity: 0.55420566
+Text: Apple Inc. has revolutionized the tech industry with its sleek
+designs and user-friendly interfaces.
+______________________________________________________________________
+Source Node 3/4
+Node ID: vec4
+Similarity: 0.3172258
+Text: An apple a day keeps the doctor away, as the saying goes.
+______________________________________________________________________
+Source Node 4/4
+Node ID: vec0
+Similarity: 0.25139993
+Text: Apple is a popular fruit known for its sweetness and crisp
+texture.
+
+```

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/llama_index/postprocessor/pinecone_native_rerank/BUILD

@@ -0,0 +1 @@
+python_sources()

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/llama_index/postprocessor/pinecone_native_rerank/__init__.py

@@ -0,0 +1,4 @@
+from llama_index.postprocessor.pinecone_native_rerank.base import PineconeNativeRerank
+
+
+__all__ = ["PineconeNativeRerank"]

llama-index-integrations/postprocessor/llama-index-postprocessor-pinecone-native-rerank/llama_index/postprocessor/pinecone_native_rerank/base.py

@@ -0,0 +1,105 @@
+import os
+from typing import List, Optional, Literal
+from llama_index.core.bridge.pydantic import Field, PrivateAttr
+from llama_index.core.callbacks import CBEventType, EventPayload
+from llama_index.core.instrumentation import get_dispatcher
+from llama_index.core.instrumentation.events.rerank import (
+    ReRankEndEvent,
+    ReRankStartEvent,
+)
+from llama_index.core.postprocessor.types import BaseNodePostprocessor
+from llama_index.core.schema import NodeWithScore, QueryBundle, MetadataMode
+
+dispatcher = get_dispatcher(__name__)
+
+
+# as of today 2024/12/7
+# pinecone itself supports 3 rerank models out of its sdk
+# https://app.pinecone.io/organizations/-Nn577_974iRsvC6nVxg/projects/a4fe57a4-b1cc-4a99-bf1d-c35a595cae4a/models
+class PineconeNativeRerank(BaseNodePostprocessor):
+    model: Literal[
+        "bge-reranker-v2-m3", "cohere-rerank-3.5", "pinecone-rerank-v0"
+    ] = Field(
+        description="supported Pinecone inference rerank model name",
+        default="bge-reranker-v2-m3",
+    )
+    top_n: int = Field(description="Top N nodes to return")
+
+    _pc: any = PrivateAttr()
+
+    def __init__(
+        self,
+        top_n: int = 2,
+        model: str = "bge-reranker-v2-m3",
+        api_key: Optional[str] = None,
+    ):
+        super().__init__(top_n=top_n, model=model)
+        try:
+            api_key = api_key or os.environ["PINECONE_API_KEY"]
+        except IndexError:
+            raise ValueError(
+                "Must pass in pinecone api key or "
+                "specify via PINECONE_API_KEY environment variable "
+            )
+
+        try:
+            from pinecone import Pinecone
+        except ImportError:
+            raise ImportError(
+                "Cannot import pinecone package, please `pip install pinecone-client`."
+            )
+
+        self._pc = Pinecone(api_key=api_key)
+
+    @classmethod
+    def class_name(cls) -> str:
+        return "PineconeNativeRerank"
+
+    def _postprocess_nodes(
+        self,
+        nodes: List[NodeWithScore],
+        query_bundle: Optional[QueryBundle] = None,
+    ) -> List[NodeWithScore]:
+        dispatcher.event(
+            ReRankStartEvent(
+                query=query_bundle, nodes=nodes, top_n=self.top_n, model_name=self.model
+            )
+        )
+
+        if query_bundle is None:
+            raise ValueError("Missing query bundle in extra info.")
+        if len(nodes) == 0:
+            return []
+
+        with self.callback_manager.event(
+            CBEventType.RERANKING,
+            payload={
+                EventPayload.NODES: nodes,
+                EventPayload.MODEL_NAME: self.model,
+                EventPayload.QUERY_STR: query_bundle.query_str,
+                EventPayload.TOP_K: self.top_n,
+            },
+        ) as event:
+            texts = [
+                node.node.get_content(metadata_mode=MetadataMode.EMBED)
+                for node in nodes
+            ]
+
+            reranked_result = self._pc.inference.rerank(
+                model=self.model,
+                top_n=self.top_n,
+                query=query_bundle.query_str,
+                documents=texts,
+                return_documents=True,
+            )
+            new_nodes = []
+            for result in reranked_result.data:
+                new_nodes.append(
+                    NodeWithScore(
+                        node=nodes[result.index].node,
+                        score=result.score,
+                    )
+                )
+            event.on_end(payload={EventPayload.NODES: new_nodes})
+        dispatcher.event(ReRankEndEvent(nodes=new_nodes))
+        return new_nodes