2. Tensor Program Abstraction
Quick search
code
Show Source
Course GitHub 中文版
Machine Learing Compiler
Table Of Contents
  • 1. Introduction
  • 2. Tensor Program Abstraction
    • 2.1. Primitive Tensor Function
    • 2.2. Tensor Program Abstraction
    • 2.3. Summary
    • 2.4. TensorIR: Tensor Program Abstraction Case Study
    • 2.5. Exercises for TensorIR
  • 3. End to End Model Execution
  • 4. Automatic Program Optimization
  • 5. Integration with Machine Learning Frameworks
  • 6. GPU and Hardware Acceleration
    • 6.1. Part 1
    • 6.2. Part 2
  • 7. Computational Graph Optimization
Machine Learing Compiler
Table Of Contents
  • 1. Introduction
  • 2. Tensor Program Abstraction
    • 2.1. Primitive Tensor Function
    • 2.2. Tensor Program Abstraction
    • 2.3. Summary
    • 2.4. TensorIR: Tensor Program Abstraction Case Study
    • 2.5. Exercises for TensorIR
  • 3. End to End Model Execution
  • 4. Automatic Program Optimization
  • 5. Integration with Machine Learning Frameworks
  • 6. GPU and Hardware Acceleration
    • 6.1. Part 1
    • 6.2. Part 2
  • 7. Computational Graph Optimization

2. Tensor Program Abstraction¶

In this chapter, we will discuss the abstractions for a single “unit” step of computation and possible MLC transformations in these abstractions.

  • 2.1. Primitive Tensor Function
  • 2.2. Tensor Program Abstraction
    • 2.2.1. Extra Structure in Tensor Program Abstraction
  • 2.3. Summary
  • 2.4. TensorIR: Tensor Program Abstraction Case Study
    • 2.4.1. Install Packages
    • 2.4.2. Prelude
    • 2.4.3. Learning one Tensor Program Abstraction – TensorIR
    • 2.4.4. Transformation
    • 2.4.5. Build and Run
    • 2.4.6. Ways to Create and Interact with TensorIR
    • 2.4.7. TensorIR Functions as Results of Transformations
    • 2.4.8. Discussions
    • 2.4.9. Summary
  • 2.5. Exercises for TensorIR
    • 2.5.1. Section 1: How to Write TensorIR
    • 2.5.2. Section 2: How to Transform TensorIR
Previous
1. Introduction
Next
2.1. Primitive Tensor Function