llama-index-networks package (run-llama#11413)

* lint

* pants tailor

* remove unnecessary files/folders

* make api_key an optional field

* make secret field optional as well

* top level imports

* prerelease version bump

* remove misleading env variables

* override gitignore to add data to demo

* add note in demo README that you can use  instead of

* update demo contributor poetry.locks pyproject.toml

* published v0.1.0; updated demo deps

llama-index-networks/.gitignore

@@ -0,0 +1,157 @@
+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
+.env*
+
+# 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
+
+# data
+data

llama-index-networks/BUILD

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

llama-index-networks/CHANGELOG.md

@@ -0,0 +1,5 @@
+# CHANGELOG
+
+## [0.1.0] - 2024-02-27
+
+- Initial release

llama-index-networks/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-networks/README.md

@@ -0,0 +1,98 @@
+# LlamaIndex Networks
+
+The `llama-index-networks` library extension allows for the creation of
+networks knowledge that can be queried and used for LLM context augmentation.
+In such networks, we have "data suppliers" and "data consumers" that participate
+and benefit from the network, respectively.
+
+## Data suppliers
+
+With this extension, you can easily wrap your llama-index `QueryEngine` with a
+`ContributorService` that exposes the query engine behind a REST (FastAPI)
+service, which renders it ready to contribute to LlamaIndex Network!
+
+```python
+from llama_index.networks.contributor import (
+    ContributorServiceSettings,
+    ContributorService,
+)
+
+query_engine = ...  # build a query engine as typically done with llama-index
+
+settings = ContributorServiceSettings()
+service = ContributorService(config=settings, query_engine=query_engine)
+app = service.app
+
+
+if __name__ == "__main__":
+    import uvicorn
+
+    uvicorn.run(app, port=8000)
+```
+
+## Data consumers
+
+With `llama-index-networks`, you can build a `NetworkQueryEngine` that is able
+to connect to set of (network) `ContributorService`'s.
+
+```python
+from llama_index.networks.contributor import ContributorClient
+from llama_index.networks.query_engine import NetworkQueryEngine
+from llama_index.llms.groq import Groq
+
+# Use ContributorClient to connect to a ContributorService
+client = ContributorClient.from_config_file(env_file=".env.contributor.client")
+contributors = [client]
+
+# NetworkQueryEngine
+llm = Groq()
+network_query_engine = NetworkQueryEngine.from_args(
+    contributors=contributors, llm=llm
+)
+
+# Can query it like any other query engine
+network_query_engine.query("Why is the sky blue?")
+```
+
+The `.env.contributor.client` file contains the parameters to connect to the
+`ContributorService` (i.e., `api_url` and `api_token`.)
+
+## Examples
+
+For a full demo, checkout the `examples/demo` sub-folder.
+
+### FAQ
+
+#### 1. Can I add my own custom endpoints to a `ContributorService`?
+
+Yes, the (Fastapi) app is accessible via `ContributorService.app` and can be
+modified to your needs.
+
+```python
+settings = ContributorServiceSettings()
+service = ContributorService(config=settings, query_engine=query_engine)
+app = service.app
+
+
+@app.get("<custom-path>")
+def custom_endpoint_logic():
+    ...
+```
+
+#### 2. How can I add authentication to a `ContributorService`?
+
+Currently, the client supports authentication through request headers as well as
+as via an `api_token` str. On the server side, you should be able to modify the
+app to your needs:
+
+```python
+settings = ContributorServiceSettings()
+service = ContributorService(config=settings, query_engine=query_engine)
+app = service.app
+
+# modify app here to include your own security mechanisms
+```
+
+As we continue to build out this library extension, we welcome any feedback or
+suggestions on what can be incorporated on this regard and others in order to
+make our tools more helpful to our users.

llama-index-networks/examples/demo/README.md

@@ -0,0 +1,97 @@
+# LlamaIndex Networks Demo
+
+In this demo, we have 3 network contributor services, whose Python source
+codes can be found in the folders listed below:
+
+- contributor-1/
+- contributor-2/
+- contributor-3/
+
+Once all of these services are up and running (see usage instructions
+attached below), then we can connect to them using a `NetworkQueryEngine`.
+The demo for doing that is found in the notebook: `network_query_engine.ipynb`.
+
+## Building and Running the 3 Network Contributor Services
+
+### Setup Environment Variables
+
+Each of the three Contributor Services wrap `QueryEngine`'s that utilize
+OpenAI as the LLM provider. As such, you'll need to supply an `OPENAI_API_KEY`.
+
+To do so, we make use of .env files. Each contributor folder requires a filled
+in `.env.contributor.service` file. You can use the `template.env.contributor.service`,
+fill in your openai-api-key and then save it as `.env.contributor.service` (you can also save it simply as `.env` as the `ContributorServiceSettings`
+class will look for `.env` file if it can't find `.env.contributor.service`).
+
+### Installation
+
+_Requires Poetry v. 1.4.1 to be installed._
+
+```sh
+cd contributor-1 && poetry install && cd -
+cd contributor-2 && poetry install && cd -
+cd contributor-3 && poetry install && cd -
+```
+
+### Running the contributor servers locally
+
+_Requires Docker to be installed._
+
+We've simplified running all three services with the help of
+`docker-compose`. It should be noted that in a real-world scenario, these
+contributor services are likely independently stood up.
+
+```sh
+docker-compose up --build
+```
+
+Any code changes will be reflected in the running server automatically without having to rebuild/restart the server.
+
+### Viewing the SWAGGER docs (local)
+
+Once the server is running locally, the standard API docs of any of
+the three contributor services can be viewed in a browser.
+
+Use any port number `{8001,8002,8003}`:
+
+```sh
+# visit in any browser
+http://localhost:<PORT>/docs#/
+```
+
+## Building and Running the NetworkQueryEngine
+
+We begin by creating a fresh environment:
+
+```sh
+pyenv virtualenv networks-demo
+pyenv activate networks-demo
+pip install jupyterlab ipykernel
+pip install llama-index llama-index-networks --pre
+```
+
+Let's create the kernel for our notebook so that we can use our newly
+created `networks-demo` virtual environment.
+
+```sh
+ipython kernel install --user --name=networks-demo
+```
+
+### Setting up the environment files
+
+As in setting up the `ContributorService`'s we need to pass in the settings
+for the `ContributorClient`'s (that communicate with their respective services).
+Simply rename the template files in `client-env-files` directory by dropping
+the term "template" in all of the .env files (e.g.,
+`template.env.contributor_1.client` becomes `.env.contributor_1.client`).
+
+### Running the Notebook
+
+Note this notebook uses OpenAI LLMs. We export the `OPENAI_API_KEY`
+at the same time that we spin up the notebook:
+
+```sh
+export OPENAI_API_KEY=<openai-api-key> && jupyter lab
+```
+
+From there you can open up the `network_query_engine.ipynb`.

llama-index-networks/examples/demo/client-env-files/template.env.contributor_1.client

@@ -0,0 +1 @@
+API_URL=http://0.0.0.0:8001

llama-index-networks/examples/demo/client-env-files/template.env.contributor_2.client

@@ -0,0 +1 @@
+API_URL=http://0.0.0.0:8002

llama-index-networks/examples/demo/client-env-files/template.env.contributor_3.client

@@ -0,0 +1 @@
+API_URL=http://0.0.0.0:8003

llama-index-networks/examples/demo/contributor-1/BUILD

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