Making Neural Networks Run in Browser With ONNX
Explore ONNX format for deploying machine learning models to edge devices, including training, conversion, and deployment using high-performance JavaScript runtime.
Extending PyTorch for Custom Compiler Targets - Accelerator Hardware Integration
Discover how to annotate PyTorch code with custom metadata for optimizing model inference on specialized hardware accelerators, focusing on LPU deployment and performance enhancement.
Open Source Privacy Preserving Inference in Medical Imaging
Explore privacy-preserving machine learning in healthcare, focusing on secure prediction-as-a-service and model testing. Learn about CrypTFlow, a system for compiling TensorFlow/ONNX to secure computation protocols.
MLOps: Inference ONNX Model in Azure ML Managed Endpoint AKS
Deploy ONNX models to Azure ML Managed Endpoints (AKS) for efficient inference. Learn step-by-step process using Python SDK v2 to optimize machine learning operations.
CPU Inference with ViT ONNX Model in Azure ML Managed Endpoint - AKS
Deploy a Vision Transformer model for CPU inference on Azure Kubernetes Service using ONNX format and Azure ML SDK v2. Learn to create managed endpoints in Azure ML Studio.
MLOPS: Inference ViVIT ONNX Model in Azure ML Managed Endpoint (AKS)
Aprenda a implementar un modelo ViViT en ONNX para inferencia en CPU utilizando Azure Kubernetes Services y onnxruntime con Python SDK v2.
Inference Engines and Audio - Machine Learning in Audio Applications
Explore inference pipelines in machine learning for audio applications, including common engines, their functionality, and creating custom solutions for platforms like Cmajor.
Tensor Query Processing - Accelerating Database Analytics with AI Hardware
Discover how Microsoft's framework compiles SQL to tensor abstractions, enabling efficient database analytics on GPUs and exploring the fusion of SQL with machine learning capabilities.
Converting T5 Large Models to ONNX Format and 8-bit Quantization with MLflow and Optimum
Master model optimization techniques by converting T5 Large to ONNX format and implementing 8INT quantization using Hugging Face Transformers Optimum, with MLflow integration for text summarization tasks.
MLOps with MLFlow: Evaluating and Quantizing Fine-Tuned LLM Models
Discover how to evaluate fine-tuned LLMs using MLflow, convert models to ONNX format, and implement INT8 quantization for optimized performance and deployment.