Skip to content

Commit

Permalink
Update hybridize.md
Browse files Browse the repository at this point in the history
  • Loading branch information
@astonzhang
astonzhang authored Jun 16, 2021
1 parent 550895c commit 0d575dd
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions chapter_computational-performance/hybridize.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -313,7 +313,7 @@ The model is decomposed into a (large binary) parameter file and a JSON descript
```

:begin_tab:`mxnet`
Earlier, we demonstrated that, after calling the `hybridize` function, the model is able to achieve superior computing performance and portability. Note that hybridization can affect model flexibility, in particular in terms of control flow.
Earlier, we demonstrated that, after calling the `hybridize` function, the model is able to achieve superior computing performance and portability. Note, though that hybridization can affect model flexibility, in particular in terms of control flow.

Besides, contrary to the `Block` instance, which needs to use the `forward` function, for a `HybridBlock` instance we need to use the `hybrid_forward` function.
:end_tab:
Expand Down Expand Up @@ -403,4 +403,4 @@ This is quite different from what we saw previously. All print statements, as de

:begin_tab:`tensorflow`
[Discussions](https://discuss.d2l.ai/t/2492)
:end_tab:
:end_tab:

0 comments on commit 0d575dd

Please sign in to comment.