This document is relevant for: Inf1, Inf2, Trn1, Trn1n

Inference

Inference on Inf2 & Trn1 (tensorflow-neuronx)

Tutorials (tensorflow-neuronx)

• HuggingFace Roberta-Base [html] [notebook]

• Using NEURON_RT_VISIBLE_CORES with TensorFlow Serving

Note

To use Jupyter Notebook see:

• Jupyter Notebook QuickStart

• Running Jupyter Notebook as script

API Reference Guide (tensorflow-neuronx)

• TensorFlow 2.x (tensorflow-neuronx) Tracing API

• TensorFlow Neuron (tensorflow-neuronx) Auto Multicore Replication (Experimental)

Misc (tensorflow-neuronx)

• TensorFlow Neuron (tensorflow-neuronx) Release Notes

Inference on Inf1 (tensorflow-neuron)

Tutorials (tensorflow-neuron)

Computer Vision Tutorials

• Tensorflow 1.x - OpenPose tutorial [html] [notebook]

• Tensorflow 1.x - ResNet-50 tutorial [html] [notebook]

• Tensorflow 1.x - YOLOv4 tutorial [html] [notebook]

• Tensorflow 1.x - YOLOv3 tutorial [html] [notebook]

• Tensorflow 1.x - SSD300 tutorial [html]

• Tensorflow 1.x - Keras ResNet-50 optimization tutorial [html] [notebook]

Natural Language Processing (NLP) Tutorials

• Tensorflow 1.x - Running TensorFlow BERT-Large with AWS Neuron [html]

• Tensorflow 2.x - HuggingFace Pipelines distilBERT with Tensorflow2 Neuron [html] [notebook]

Utilizing Neuron Capabilities Tutorials

• Tensorflow 1.x & 2.x - Using NEURON_RT_VISIBLE_CORES with TensorFlow Serving [html]

Note

To use Jupyter Notebook see:

• Jupyter Notebook QuickStart

• Running Jupyter Notebook as script

Additional Examples (tensorflow-neuron)

• AWS Neuron Samples GitHub Repository

API Reference Guide (tensorflow-neuron)

• TensorFlow 2.x (tensorflow-neuron) Tracing API

• TensorFlow 2.x (tensorflow-neuron) analyze_model API

• TensorFlow 1.x (tensorflow-neuron) Compilation API

• TensorFlow Neuron (tensorflow-neuron) Auto Multicore Replication (Experimental)

Misc (tensorflow-neuron)

• TensorFlow Neuron (tensorflow-neuron (TF1.x)) Release Notes

• TensorFlow Neuron (tensorflow-neuron (TF2.x)) Release Notes

• TensorFlow Neuron (tensorflow-neuron (TF2.x)) Accelerated (torch-neuron) Python APIs and Graph Ops

• TensorFlow Neuron (tensorflow-neuron (TF1.x)) Supported operators

This document is relevant for: Inf1, Inf2, Trn1, Trn1n