Low GPU usage by Keras / Tensorflow? - Stack Overflow
François Chollet on Twitter: "PlaidML is an open-source Keras backend that runs on OpenCL, with good performance relative to TensorFlow: https://t.co/tCmzXHfFUO https://t.co/3YAQI0EMX5" / Twitter
TensorFlow OpenCL | Complete Guide to TensorFlow OpenCL
Accelerate your machine learning networks using TVM and the Adreno OpenCL ML APIs on Adreno GPUs - Qualcomm Developer Network
Bringing AMDGPUs to TVM Stack and NNVM Compiler with ROCm
CitiusSynapse: A Deep Learning Framework for Embedded Systems
GitHub - kpot/avalanche: Avalanche - OpenCL deep learning framework with a backend for Keras
Implementing OpenCL backend for pytorch - hardware-backends - PyTorch Dev Discussions
OpenCL support · Issue #22 · tensorflow/tensorflow · GitHub
opencl | LearnOpenCV
python - Keras Machine Learning Code are not using GPU - Stack Overflow
Train neural networks using AMD GPU and Keras | by Mattia Varile | Towards Data Science
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
TensorFlow in Context
Train neural networks using AMD GPU and Keras | by Mattia Varile | Towards Data Science
OpenCL Overview
Accelerate your machine learning networks using TVM and the Adreno OpenCL ML APIs on Adreno GPUs - Qualcomm Developer Network
a) Memory access pattern in OpenCL standard 1.x. (b) OpenCL pipes; in... | Download Scientific Diagram