Difference between revisions of "AI Help"

Latest revision as of 19:48, 15 July 2024

New User's Guide

For new users on HiPerGator, please read Getting Started to become familiar with the HiPerGator system and take New user training with step-by-step instructions on how to use HiPerGator.

AI Education and Training provides learning materials and training videos on various AI topics. JupyterHub and Jupyter Notebooks on HiPerGator are popular platforms for developing and running AI programs.

AI Computing Resources: HiPerGator and HiPerGator AI provide world class computing infrastructure for conducting AI research and building AI applications. To purchase resource allocations, please refer to Purchase Request. For computing server features and GPU access, please visit Available Node Features and GPU Access.

AI Frameworks and Tools

General AI Frameworks

AI frameworks provide building blocks for designing and training machine learning and deep learning models. The following AI frameworks are available on HiPerGator.

PyTorch is a deep learning framework developed by Facebook AI Research Lab and has interfaces for Python, Java, and C++, which is most commonly used with Python. It supports training on both GPU and CPUs, as well as distributed training and multi-GPU models.

TensorFlow is an open-source AI framework/platform developed by the Google Brain team. Keras is an open-source neural network library that runs on top of TensorFlow. With TensorFlow 2.0, the Keras API has been integrated into TensorFlow's core library and serves as a high-level Python interface for TensorFlow. TensorFlow supports both GPU and CPUs, as well as multi-GPU and distributed training.

Tensorboard is a visualization tool for monitoring neural network training.

Sci-kit learn is a Python library for machine learning and statistical modeling. It is available in many of the Python modules on HiPerGator.

Matlab provides convenient toolboxes for machine learning, deep learning, computer vision, and automatic driving, which are supported on both CPUs and GPUs.

RAPIDS AI is a suite of CUDA-enabled open-source software libraries and APIs by NVIDIA. It accelerates end-to-end data science pipelines by providing a familiar dataframe API. It supports machine learning integration without the typical serialization costs and enables multi-node, multi-GPU deployments for faster processing of large datasets.

Fastai simplifies training fast and accurate neural nets using modern best practices. It can be used without any installation by using Google Colab.

Triton is a Nvidia Inference Server that provides a cloud and edge inferencing solution optimized for both CPUs and GPUs. Triton supports an HTTP/REST and GRPC protocol that allows remote clients to request inferencing for any model being managed by the server.

AI Software for Specific Domains

Comprehensive software libraries, toolkits, and frameworks for big data and AI research are available on HiPerGator:

Data Science Platform is the practice of using mathematics, programming, analytics, AI, and machine learning to discover valuable insights within large data sets.

Natural Language Processing page describes a collection of Natural Language Processing software on HiperGator, such as Nemo, BioNemo, Llama2, Llama3, Mistral AI, and Gemma LLMs. NLP is a part of AI that helps computers understand and respond to human language. It's used in things like voice assistants, chatbots, and translation apps.

Computer Vision page describes the software, tools, and environments on HiPerGator for image and video processing.

Healthcare and Life Sciences page describes the collection of Healthcare and Life Sciences software for medical image analysis, disease detection, and genomic data interpretation on HiperGator AI, such as AlphaFold, BioNemo, Clara Parabricks, and MONAI.

Simulation and Modeling HiPerGator supports advanced simulation tools, including NVIDIA Modulus and Omniverse, to predict the behavior of systems within a real-world scenario.

Nvidia Modulus is a neural network framework that blends the power of physics in the form of governing partial differential equations (PDEs) with data to build high-fidelity, parameterized surrogate models with near-real-time latency. Examples of running modulus on HiperGator are available on /data/ai/examples/pinn/modulus or AI_Examples.

Nvidia Omniverse provides a platform for virtual collaboration and real-time photorealistic simulation. It allows developers, designers, and engineers to collaborate in simulated environments, facilitating the creation of complex simulations that integrate physical accuracy and visual fidelity.

AI Reference Datasets

A variety of AI reference datasets are located in /data/ai/ref-data. Browse the catalog of all available AI reference datasets to learn more.

AI Examples

A suite of short examples is provided using the software stack on HiPerGator to do different tasks in various AI domains, located in /data/ai/examples. Browse the catalog of all available AI examples to learn more.

AI Benchmarks and Models

Commonly used benchmarks and models are provided on HiPerGator at /data/ai/benchmarks and /data/ai/models. Browse the catalog of all available AI models to learn more.

AI Zoom Support

You can schedule a Zoom call with an AI support staff member. For details see Remote Support

@@ Line 1: / Line 1: @@
 [[Category:Help]]
-= New User's Guide =
+{|align=right
-For new users on HiPerGator, please read [[Getting Started]] to get yourself familiar with HiPerGator system and take [[New user training]] with step-by-step instructions on how to use HiPerGator.
+  |__TOC__
+  |}
+== New User's Guide ==
+For new users on HiPerGator, please read [[Getting Started]] to become familiar with the HiPerGator system and take [[New user training]] with step-by-step instructions on how to use HiPerGator.
-[[AI Education and Training]] provides learning materials and training videos on various AI topics. [[JupyterHub]] and [[Jupyter Notebooks]] on HiPerGator are popular platforms for developing and running AI programs.
+[[AI Education and Training]] provides learning materials and training videos on various AI topics. [[Jupyter_Notebooks#JupyterHub | JupyterHub]] and [[Jupyter Notebooks]] on HiPerGator are popular platforms for developing and running AI programs.
-= AI Software =
+AI Computing Resources: [https://www.rc.ufl.edu/get-started/hipergator/ HiPerGator and HiPerGator AI] provide world class computing infrastructure for conducting AI research and building AI applications. To purchase resource allocations, please refer to [https://www.rc.ufl.edu/get-started/purchase-request/ Purchase Request]. For computing server features and GPU access, please visit [[Available Node Features]] and [[GPU Access]].
-A comprehensive software stack for AI research is available on HiPerGator for both CPU and GPU accelerated applications. The [[Nlp|NLP]] page has more info on software for NLP.
+== AI Frameworks and Tools ==
+=== General AI Frameworks===
-== AI Frameworks ==
 AI frameworks provide building blocks for designing and training machine learning and deep learning models. The following AI frameworks are available on HiPerGator.
-=== PyTorch===
+*'''[[PyTorch]]''' is a deep learning framework developed by Facebook AI Research Lab and has interfaces for Python, Java, and C++, which is most commonly used with Python. It supports training on both GPU and CPUs, as well as distributed training and multi-GPU models.
-[https://pytorch.org/ PyTorch] is a deep learning framework developed by Facebook AI Research Lab and has interfaces for Python, Java, and C++, but is most commonly used with Python. It supports training on both GPU and CPUs, as well as distributed training and multi-GPU models. See our '''[[PyTorch|PyTorch quickstart page]]''' for help getting started using PyTorch on HiPerGator.
+*'''[[TensorFlow]]''' is an open-source AI framework/platform developed by the Google Brain team. [https://keras.io/ Keras] is an open-source neural network library that runs on top of TensorFlow. With TensorFlow 2.0, the Keras API has been integrated into TensorFlow's core library and serves as a high-level Python interface for TensorFlow. TensorFlow supports both GPU and CPUs, as well as multi-GPU and distributed training.
+*''' [[Tensorboard]]''' is a visualization tool for monitoring neural network training.
+*[https://scikit-learn.org/stable/ '''Sci-kit learn'''] is a Python library for machine learning and statistical modeling. It is available in many of the [[Python]] modules on HiPerGator.
+*'''[[Matlab]]''' provides convenient toolboxes for machine learning, deep learning, computer vision, and automatic driving, which are supported on both CPUs and GPUs.
+*'''[[Rapidsai|RAPIDS AI]]''' is a suite of CUDA-enabled open-source software libraries and APIs by NVIDIA. It accelerates end-to-end data science pipelines by providing a familiar dataframe API. It supports machine learning integration without the typical serialization costs and enables multi-node, multi-GPU deployments for faster processing of large datasets.
+*'''[[Fastai]]''' simplifies training fast and accurate neural nets using modern best practices. It can be used without any installation by using [https://colab.research.google.com Google Colab].
+*'''[[TRITONSERVER|Triton]]''' is a Nvidia Inference Server that provides a cloud and edge inferencing solution optimized for both CPUs and GPUs. Triton supports an HTTP/REST and GRPC protocol that allows remote clients to request inferencing for any model being managed by the server.
+=== AI Software for Specific Domains ===
-=== Tensorflow/Keras ===
+Comprehensive software libraries, toolkits, and frameworks for big data and AI research are available on HiPerGator:
-[https://www.tensorflow.org/ TensorFlow] is an open-source AI framework/platform developed by the Google Brain team. [https://keras.io/ Keras] is an open-source neural network library which runs on top of TensorFlow. With TensorFlow 2.0, the Keras API has been integrated in TensorFlow's core library and serves as a high-level Python interface for TensorFlow. TensorFlow supports both GPU and CPUs, as well as multi-GPU and distributed training. APIs are available for Python, Java, Go and C++. See our [[TensorFlow|TensorFlow quickstart page]] for help getting started using TensorFlow on HiPerGator.
-'''[[Tensorboard]]''' is a visualization tool for monitoring neural network training.
+'''[[Data Science Platform]]''' is the practice of using mathematics, programming, analytics, AI, and machine learning to discover valuable insights within large data sets.
-=== Sci-kit Learn ===
+'''[[Nlp|Natural Language Processing]]''' page describes a collection of Natural Language Processing software on HiperGator, such as Nemo, BioNemo, Llama2, Llama3, Mistral AI, and Gemma LLMs. NLP is a part of AI that helps computers understand and respond to human language. It's used in things like voice assistants, chatbots, and translation apps.
-[https://scikit-learn.org/stable/ Sci-kit learn] is a Python library for machine learning and statistical modeling. It is available in many of the '''[[Python]]''' modules on HiPerGator.
-=== MATLAB ===
+'''[[Computer Vision]]''' page describes the software, tools, and environments on HiPerGator for image and video processing.
-'''[[Matlab]]''' provides convenient toolboxes for machine learning, deep learning, computer vision and automatic driving, which  are supported on both CPUs and GPUs.
-=== Fastai ===
+'''[[Healthcare and Life Sciences]]''' page describes the collection of Healthcare and Life Sciences software for medical image analysis, disease detection, and genomic data interpretation on HiperGator AI, such as AlphaFold, BioNemo, Clara Parabricks, and MONAI.
-[[Fastai]] simplifies training fast and accurate neural nets using modern best practices. It can be used without any installation by using [https://colab.research.google.com Google Colab].
-== NVIDIA AI Software ==
+'''Simulation and Modeling''' HiPerGator supports advanced simulation tools, including NVIDIA Modulus and Omniverse, to predict the behavior of systems within a real-world scenario.
-Nvidia provides comprehensive, GPU-accelerated software libraries, toolkits, frameworks and packages for big-data and AI applications. Many of the libraries are included in the CUDA installation on HiPerGator, such as cuDNN. The following domain specific CUDA enabled tools are available on HiPerGator:
-=== Megatron-LM ===
+*Nvidia '''[[Modulus]]''' is a neural network framework that blends the power of physics in the form of governing partial differential equations (PDEs) with data to build high-fidelity, parameterized surrogate models with near-real-time latency. Examples of running modulus on HiperGator are available on /data/ai/examples/pinn/modulus or [[AI_Examples]].
-[https://github.com/NVIDIA/Megatron-LM Megatron-LM] can train several architectures of language models, including an GPT, T5, and an improved BERT. Megatron also recently added a transformer-based image classification architecture.
-=== NeMo ===
+*Nvidia '''[[Omniverse]]''' provides a platform for virtual collaboration and real-time photorealistic simulation. It allows developers, designers, and engineers to collaborate in simulated environments, facilitating the creation of complex simulations that integrate physical accuracy and visual fidelity.
-[https://developer.nvidia.com/nvidia-nemo Nvidia NeMo] is an open-source Python, GPU-accelerated toolkit for conversational AI, including speech recognition (ASR), natural language processing (NLP) and text to speech (TTS) applications. NeMo is available via a container in the apps folder.
-=== Clara Parabricks ===
+==AI Reference Datasets==
-[[Parabricks|Nvidia Clara Parabricks]] ([https://developer.nvidia.com/Clara-parabricks]) is a computation framework for genomics applications. It builds GPU accelerated libraries, pipelines, and reference AI workflows for genomics research. We have a license for this software through 2021.
+A variety of AI reference datasets are located in <code>/data/ai/ref-data</code>. Browse the '''[[AI Reference Datasets|catalog of all available AI reference datasets]]''' to learn more.
-=== Clara MONAI ===
+==AI Examples ==
-[[Monai|Nvidia Clara MONAI]] ([https://monai.io]) is the open-source foundation being created by Project MONAI. MONAI is a freely available, community-supported, PyTorch-based framework for deep learning in healthcare imaging. It provides domain-optimized foundational capabilities for developing healthcare imaging training workflows in a native PyTorch paradigm.
+A suite of short examples is provided using the software stack on HiPerGator to do different tasks in various AI domains, located in <code>/data/ai/examples</code>. Browse the '''[[AI Examples|catalog of all available AI examples]]''' to learn more.
-=== Modulus ===
+== AI Benchmarks and Models ==
-[[Modulus|Nvidia Modulus]] ([https://developer.nvidia.com/modulus]) is a neural network framework that blends the power of physics in the form of governing partial differential equations (PDEs) with data to build high-fidelity, parameterized surrogate models with near-real-time latency.
+Commonly used benchmarks and models are provided on HiPerGator at <code>/data/ai/benchmarks</code> and <code>/data/ai/models</code>. Browse the '''[[AI Models|catalog of all available AI models]]''' to learn more.
-=AI Reference Datasets=
+== AI Zoom Support ==
-A variety of reference machine learning and AI datasets are located in <code>/data/ai</code>. Browse the [[AI Reference Datasets|catalog of all available AI reference datasets]] to learn more.
+You can schedule a Zoom call with an AI support staff member. For details see [https://help.rc.ufl.edu/doc/Remote_Support Remote Support]