diff --git a/community/README.md b/community/README.md index 52deb5f33..7197cddb8 100644 --- a/community/README.md +++ b/community/README.md @@ -37,7 +37,7 @@ Community examples are sample code and deployments for RAG pipelines that are no * [NVIDIA RAG Streaming Document Ingestion Pipeline](./streaming_ingest_rag) - This example demonstrate the construction of a performance-oriented pipeline that accepts a stream of heterogenous documents, divides the documents into smaller segments or chunks, computes the embedding vector for each of these chunks, and uploads the text chunks along with their associated embeddings to a Vector Database. This pipeline builds on the [Morpheus SDK](https://docs.nvidia.com/morpheus/index.html) to take advantage of end-to-end asynchronous processing. This pipeline showcases pipeline parallelism (including CPU and GPU-accelerated nodes), as well as, a mechanism to horizontally scale out data ingestion workers. + This example demonstrates the construction of a performance-oriented pipeline that accepts a stream of heterogenous documents, divides the documents into smaller segments or chunks, computes the embedding vector for each of these chunks, and uploads the text chunks along with their associated embeddings to a Vector Database. This pipeline builds on the [Morpheus SDK](https://docs.nvidia.com/morpheus/index.html) to take advantage of end-to-end asynchronous processing. This pipeline showcases pipeline parallelism (including CPU- and GPU-accelerated nodes) as well as a mechanism to horizontally scale out data ingestion workers. * [NVIDIA Live FM Radio ASR RAG](./fm-asr-streaming-rag) @@ -45,7 +45,7 @@ Community examples are sample code and deployments for RAG pipelines that are no * [NVIDIA ORAN chatbot multimodal Assistant](./oran-chatbot-multimodal/) - This example is designed to make it extremely easy to set up your own retrieval-augmented generation chatbot for ORAN techncial specifications and processes. The backend here calls the NVIDIA NeMo Service, which makes it very easy to deploy on a thin client or Virtual Machine (ie, without a GPU setup). + This example is designed to make it extremely easy to set up your own retrieval-augmented generation chatbot for ORAN technical specifications and processes. The backend here calls the NVIDIA NeMo Service, which makes it very easy to deploy on a thin client or Virtual Machine (i.e., without a GPU setup). * [NVIDIA Retrieval Customization](./synthetic-data-retriever-customization/) @@ -61,7 +61,7 @@ Community examples are sample code and deployments for RAG pipelines that are no * [NVIDIA Event Driven RAG for CVE Analysis with NVIDIA Morpheus](./event-driven-rag-cve-analysis/) - This example demonstrates how NVIDIA Morpheus, NIM microservices, and RAG pipelines can be integrated to create LLM-based agent pipelines. These pipelines will be used to automatically and scalably traige and detect Common Vulnerabilities and Exposures (CVEs) in Docker containers using references to source code, dependencies, and information about the CVEs. + This example demonstrates how NVIDIA Morpheus, NIM microservices, and RAG pipelines can be integrated to create LLM-based agent pipelines. These pipelines will be used to automatically and scalably triage and detect Common Vulnerabilities and Exposures (CVEs) in Docker containers using references to source code, dependencies, and information about the CVEs. * [Digital Human Security Analyst with NVIDIA Morpheus](./digital-human-security-analyst/) diff --git a/community/autonomous_5g_slicing_lab/README.md b/community/autonomous_5g_slicing_lab/README.md index 501ebc078..ed149cc43 100644 --- a/community/autonomous_5g_slicing_lab/README.md +++ b/community/autonomous_5g_slicing_lab/README.md @@ -15,7 +15,7 @@ A shutdown notebook is also provided to reinitialize the lab if needed. To run the lab, ensure you have: - **Operating System**: Ubuntu (tested on 20.04 or later). - **Hardware**: - - CPU: 12+ cores @ 3,8 GHz, AVX-512 is a necessary + - CPU: 12+ cores @ 3.8 GHz, AVX-512 is required - RAM: 32 GB - OS: Modern Linux (e.g., Ubuntu 22.04) - Docker & Docker Compose (latest stable) @@ -28,7 +28,7 @@ To run the lab, ensure you have: ## Lab Setup -The lab setup configures a fully functional 5G network simulation environment. To set up the lab, first you will need to run autonomous_5g_slicing_lab/Automatic_5G_Network_Lab_Setup.ipynb to configure your environment, and then you will need to run the Jupyter notebook located at `autonomous_5g_slicing_lab/llm-slicing-5g-lab/DLI_Lab_Setup.ipynb`. The notebook automates the following steps: +The lab setup configures a fully functional 5G network simulation environment. To set up the lab, first you will need to run `autonomous_5g_slicing_lab/Automatic_5G_Network_Lab_Setup.ipynb` to configure your environment, and then you will need to run the Jupyter notebook located at `autonomous_5g_slicing_lab/llm-slicing-5g-lab/DLI_Lab_Setup.ipynb`. The notebook automates the following steps: 1. **Install Dependencies**: Installs iPerf3 and Python packages required for the lab, then restarts the Jupyter kernel to apply changes. 2. **Compile FlexRIC and gNodeB**: Builds the FlexRIC and gNodeB components using a provided script, preparing the RAN Intelligent Controller and base station software. @@ -41,8 +41,8 @@ The lab setup configures a fully functional 5G network simulation environment. T 9. **Generate Traffic and Log Data**: Runs iPerf clients on the UEs to generate UDP traffic at alternating speeds (30 Mbps and 120 Mbps), logging performance metrics (e.g., bitrate, packet loss) to a Kinetica database and local log files. In summary, to start your lab, you need to follow these steps: -1. **Open 'autonomous_5g_slicing_lab/Automatic_5G_Network_Lab_Setup.ipynb' in the main directory and set up your environment keys -2. **Open 'autonomous_5g_slicing_lab/llm-slicing-5g-lab/DLI_Lab_Setup.ipynb and set up your 5G Network Environment +1. Open `autonomous_5g_slicing_lab/Automatic_5G_Network_Lab_Setup.ipynb` in the main directory and set up your environment keys. +2. Open `autonomous_5g_slicing_lab/llm-slicing-5g-lab/DLI_Lab_Setup.ipynb` and set up your 5G Network Environment. ## Running the Agentic Workflow @@ -102,13 +102,13 @@ For issues or contributions, please open a GitHub issue or submit a pull request 1. [Aaron Bossert](https://www.linkedin.com/in/aaron-bossert/), Director of Solutions Engineering, [Kinetica](https://www.kinetica.com/) 2. [Stefan Spettel](https://www.linkedin.com/in/stefan-spettel/), CEO & Co-Founder, [phine.tech](https://phine.tech/) -4. [Fransiscus Asisi Bimo](https://www.linkedin.com/in/fransiscusbimo/), Ph.D., National Taiwan University of Science and Technology -6. [Shibani Likhite](https://www.linkedin.com/in/shibani-likhite/), Solution Architect, NVIDIA -7. [Swastika Dutta](https://www.linkedin.com/in/swastika-dutta/), Solution Architect, NVIDIA -8. [Ari Uskudar](https://www.linkedin.com/in/ari-u-628b30148/), Product Manager, NVIDIA. -9. [Maria Amparo Canaveras Galdon](https://www.linkedin.com/in/amparo-canaveras-b2152522/), Senior Solution Architect, NVIDIA -10. [Ira Bargon III](https://www.linkedin.com/in/ira-bargon-iii-a8661514/), Sr. Director of Technology and Innovation, Sterling -11. [Lukas Rothender](https://www.linkedin.com/in/lukas-rotheneder-82984327a/), CTO & Co-Founder, [phine.tech](https://phine.tech/) +3. [Fransiscus Asisi Bimo](https://www.linkedin.com/in/fransiscusbimo/), Ph.D., National Taiwan University of Science and Technology +4. [Shibani Likhite](https://www.linkedin.com/in/shibani-likhite/), Solution Architect, NVIDIA +5. [Swastika Dutta](https://www.linkedin.com/in/swastika-dutta/), Solution Architect, NVIDIA +6. [Ari Uskudar](https://www.linkedin.com/in/ari-u-628b30148/), Product Manager, NVIDIA. +7. [Maria Amparo Canaveras Galdon](https://www.linkedin.com/in/amparo-canaveras-b2152522/), Senior Solution Architect, NVIDIA +8. [Ira Bargon III](https://www.linkedin.com/in/ira-bargon-iii-a8661514/), Sr. Director of Technology and Innovation, Sterling +9. [Lukas Rothender](https://www.linkedin.com/in/lukas-rotheneder-82984327a/), CTO & Co-Founder, [phine.tech](https://phine.tech/) diff --git a/docs/architecture.md b/docs/architecture.md index dff36ea4d..7a49e3a0a 100644 --- a/docs/architecture.md +++ b/docs/architecture.md @@ -36,7 +36,7 @@ The default sample deployment contains: - A Chain Server uses [LangChain](https://github.com/langchain-ai/langchain/) and [LlamaIndex](https://www.llamaindex.ai/) for combining language model components and easily constructing question-answering from a company's database. -- [Sample Jupyter Notebooks](jupyter-server.md) and [](./frontend.md) so that you can test the chat system in an interactive manner. +- [Sample Jupyter Notebooks](https://nvidia.github.io/GenerativeAIExamples/0.7.0/jupyter-server.html) and [RAG Playground Web Application](./frontend.md) so that you can test the chat system in an interactive manner. - [Milvus](https://milvus.io/docs/install_standalone-docker.md) or [pgvector](https://github.com/pgvector/pgvector) - Embeddings are stored in a vector database. Milvus is an open-source vector database capable of NVIDIA GPU-accelerated vector searches. diff --git a/docs/chain-server.md b/docs/chain-server.md index 84a2c2487..c9886708f 100644 --- a/docs/chain-server.md +++ b/docs/chain-server.md @@ -78,7 +78,7 @@ As a simple example, consider the following steps that show how to add support f docker compose up -d --build ``` -After the containers start, ingest a Juypter Notebook to the knowledge base and then query the LLM about the notebook. +After the containers start, ingest a Jupyter Notebook to the knowledge base and then query the LLM about the notebook. ## Chain Server REST API Reference diff --git a/docs/change-model.md b/docs/change-model.md index ef9d78d04..a3d5bf339 100644 --- a/docs/change-model.md +++ b/docs/change-model.md @@ -48,7 +48,7 @@ You can determine the available model names using one of the following methods: View the sample Python code and get the model name from the `model` argument to the `client.embeddings.create` method. - Install the [langchain-nvidia-ai-endpoints](https://pypi.org/project/langchain-nvidia-ai-endpoints/) Python package from PyPi. - Use the `get_available_models()` method to on an instance of an `NVIDIAEmbeddings` object to list the models. + Use the `get_available_models()` method on an instance of an `NVIDIAEmbeddings` object to list the models. Refer to the package web page for sample code to list the models. diff --git a/docs/frontend.md b/docs/frontend.md index df6bf2ff0..8953aed30 100644 --- a/docs/frontend.md +++ b/docs/frontend.md @@ -21,7 +21,7 @@ The web application provides a user interface to the RAG [chain server](./chain- - By selecting **Use knowledge base**, the chat bot returns responses that are augmented with data from documents that you uploaded and were stored in the vector database. - To store content in the vector database, click **Knowledge Base** in the upper right corner and upload documents. -![Diagram](images/image4.jpg) +![NVIDIA LLM Playground converse interface showing a llama-2-70b-chat window with a user prompt, assistant reply, an input box, and control buttons.](images/image4.jpg) ## Web Application Design