frame
Accelerator analyzer
A tool for analyzing and optimizing DNN accelerators
FRAME: Fast Roofline Analytical Modeling and Estimation
31 stars
2 watching
6 forks
Language: Jupyter Notebook
last commit: over 1 year ago
Linked from 1 awesome list
acceleratorattention-modelcost-modeldnn
Related projects:
Repository | Description | Stars |
---|---|---|
| An analytical cost model for evaluating the performance and energy efficiency of different dataflows in deep learning accelerators | 186 |
| A platform for exploring and evaluating deep learning hardware acceleration using a full-system simulation approach | 828 |
| An autonomous framework for optimizing hardware mapping of deep neural networks on accelerators | 35 |
| Provides tools and frameworks for generating and reconstructing compressed sensing accelerated MRI sequences with motion correction capabilities. | 112 |
| An investigation tool to analyze the performance impact of inline optimizations in a .NET runtime environment | 15 |
| An autoML tool that searches and optimizes neural network architectures using tensorflow and keras. | 30 |
| A modular data movement accelerator designed to support various platforms and protocols | 102 |
| A framework that accelerates deep neural networks in web browsers using optimized models and GPU acceleration. | 1,978 |
| A tool to streamline fine-tuning of multimodal models for vision-language tasks | 1,415 |
| A .NET library that enables developers to run complex applications on various hardware platforms without writing additional C kernel code. | 177 |
| An extension for Visual Studio to scaffold new .NET MAUI app projects quickly and efficiently | 365 |
| A toolbox for efficient hyperparameter tuning in deep learning using Bayesian optimization and automatic differentiation | 23 |
| A platform for analyzing and managing binary assemblies using machine learning and data mining techniques | 622 |
| A set of tools for analyzing and debugging iOS app performance issues | 1,486 |
| An extension for Directus that provides AI-powered media analysis capabilities. | 63 |