Skip to content

Commit

Permalink
Update accelerate_pytorch.md
Browse files Browse the repository at this point in the history
minor format updates
  • Loading branch information
writinwaters authored Sep 20, 2022
1 parent a594815 commit d727a8d
Showing 1 changed file with 6 additions and 6 deletions.
12 changes: 6 additions & 6 deletions docs/lang/articles/get-started/accelerate_pytorch.md
Original file line number Diff line number Diff line change
Expand Up @@ -119,12 +119,12 @@ The function of the depth wise convolution operator:

The following subsections take the Baseline implementations as an example to show you how to implement a depthwise convolution operator with Python, PyTorch, CUDA, and Taichi, and how they compare to each other. With Taichi, you can accelerate your ML model development with ease and get rid of the tedious low-level parallel programming.

| Implementation | Readability | Performance |
| -------------- | ----------- | ----------------------------------------- |
| Python | Excellent | The slowest |
| PyTorch | Poor | Slow |
| CUDA | Poor | Fast |
| Taichi | Excellent | Comparable to that of CUDA or even better |
| Implementation | Readability | Performance |
| :-------------- | :----------- | :----------------------------------------- |
| Python | Excellent | The slowest |
| PyTorch | Poor | Slow |
| CUDA | Poor | Fast |
| Taichi | Excellent | Comparable to that of CUDA or even better |

### Implement a depthwise convolution operator with Python

Expand Down

0 comments on commit d727a8d

Please sign in to comment.