[Relay] Modify create_executor to pass params #8418
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mikepapadim
force-pushed
the
executor_api_fix
branch
2 times, most recently
from
2a937d8 to
9bc6874
Compare
YuchenJin
approved these changes
Jul 7, 2021
Contributor
YuchenJin
left a comment
YuchenJin
left a comment
There was a problem hiding this comment.
LGTM! Thanks @mikepapadim for the contribution!
YuchenJin
reviewed
Jul 7, 2021
| device = _nd.device(str(target), 0) | ||
|
|
||
| if params is not None: | ||
| mod = IRModule.from_expr(bind_params_by_name(mod["main"], params)) |
Contributor
There was a problem hiding this comment.
Do we need to consider what if the mod does not contain a "main" function, or the mod contains multiple functions(subgraphs), each with a different params dict?
Contributor
There was a problem hiding this comment.
I think it's fine to throw an error if cannot find "main" in the Module, so ignore this comment. :)
Contributor
Author
There was a problem hiding this comment.
Yes, and it is passing all the CI stages.
mikepapadim
force-pushed
the
executor_api_fix
branch
from
9bc6874 to
fd4effd
Compare
ZihengJiang
approved these changes
Jul 13, 2021
Contributor
|
Merged now. Thanks @mikepapadim |
ylc
pushed a commit
to ylc/tvm
that referenced
this pull request
Sep 29, 2021
* Overload create_executor to accept params * [fix] Add stringdoc for new param in create_executor
zxy844288792
pushed a commit
to zxy844288792/tvm
that referenced
this pull request
Mar 4, 2022
* Overload create_executor to accept params * [fix] Add stringdoc for new param in create_executor
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Following the forum discussion (https://discuss.tvm.apache.org/t/questions-about-tvm-executors-and-its-apis/10289/3) and the tutorials https://tvm.apache.org/docs/tutorials/frontend/from_keras.html#sphx-glr-tutorials-frontend-from-keras-py and https://tvm.apache.org/docs/tutorials/frontend/from_onnx.html#compile-the-model-with-relay there is some performance mismatch between the two apis.
This repo captures the performance issue.
It seems when one uses the relay API as in the example below,
paramsare not passed properly to the TIR and all optimizations regarding constants (i.e., constant folding) are prevented. Therefore, this leads to a performance mismatch with therelay.vm.compile(mod, target=target, params=params).Example usage:
This PR modifies the above to accept:
With this modification
relay.vm.compileandrelay.create_executorgenerate the same bytecodes while applying the same opts.@jroesch @tqchen @YuchenJin @ZihengJiang