How to
Tutorials
References
Deep Dive
MISC
Introduction¶
Installing TVM¶
Compiling and Optimizing a Model with TVMC¶
Compiling and Optimizing a Model with the Python AutoScheduler¶
Working with Operators Using Tensor Expressions¶
Optimizing Operators with Templates and AutoTVM¶
Cross Compilation and RPC¶
Quick Start Tutorial for Compiling Deep Learning Models¶
Optimizing Operators with Auto-scheduling¶
Compile PyTorch Models¶
Compile Tensorflow Models¶
Compile MXNet Models¶
Compile ONNX Models¶
Compile Keras Models¶
Compile TFLite Models¶
Compile CoreML Models¶
Compile YOLO-V2 and YOLO-V3 in DarkNet Models¶
Compile Caffe2 Models¶
Building a Graph Convolutional Network¶
Deploy the Pretrained Model on Android¶
Deploy the Pretrained Model on Raspberry Pi¶
Compile PyTorch Object Detection Models¶
Deploy a Framework-prequantized Model with TVM¶
Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)¶
Deploy a Quantized Model on Cuda¶
Deploy a Hugging Face Pruned Model on CPU¶
Deploy Single Shot Multibox Detector(SSD) model¶
Using External Libraries in Relay¶
Schedule Primitives in TVM¶
Intrinsics and Math Functions¶
Scan and Recurrent Kernel¶
External Tensor Functions¶
Use Tensorize to Leverage Hardware Intrinsics¶
Compute and Reduce with Tuple Inputs¶
Use Tensor Expression Debug Display (TEDD) for Visualization¶
Reduction¶
How to optimize GEMM on CPU¶
How to optimize convolution on GPU¶
How to optimize convolution using TensorCores¶
How to optimize matmul with Auto TensorCore CodeGen¶
Tuning High Performance Convolution on NVIDIA GPUs¶
Auto-tuning a Convolutional Network for NVIDIA GPU¶
Auto-tuning a Convolutional Network for x86 CPU¶
Auto-tuning a Convolutional Network for ARM CPU¶
Auto-tuning a Convolutional Network for Mobile GPU¶
Auto-scheduling a Convolution Layer for GPU¶
Auto-scheduling a Neural Network for x86 CPU¶
Auto-scheduling a Neural Network for NVIDIA GPU¶
Auto-scheduling a Neural Network for ARM CPU¶
Auto-scheduling a Neural Network for mali GPU¶
Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule¶
Writing a Customized Pass¶
How to Use TVM Pass Infra¶
Bring Your Own Datatypes to TVM¶
Introduction to TOPI¶
microTVM Reference Virtual Machines¶
microTVM with TFLite Models¶
Gallery generated by Sphinx-Gallery