Skip to content

Conversation

@everton1984
Copy link
Contributor

Use the shape argument type instead of a fixed Int32 type, this expects the shape type to always be an integer. Fix #11794 .

@gromero
Copy link
Contributor

gromero commented Jul 13, 2022

@tqchen @jroesch cc

@masahi
Copy link
Member

masahi commented Jul 13, 2022

LGTM, can you add the test from #11794?

@everton1984
Copy link
Contributor Author

@masahi Sure, I just added it under test_type_infer.py which seemed reasonable to me but please let me know if you believe it should be somewhere else.

@everton1984 everton1984 force-pushed the argmax_return_type branch 4 times, most recently from c0c5ecb to 274a1e6 Compare July 14, 2022 13:51
Fix ArgReduce automatic return type inference by forcing it to use the
datatype of the shape of the Tensor instead of the fixed Int32.

Including additional tests.
Copy link
Member

@junrushao junrushao left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@junrushao junrushao merged commit b9fa576 into apache:main Jul 14, 2022
xinetzone pushed a commit to daobook/tvm that referenced this pull request Nov 25, 2022
…e#12083)

Fix ArgReduce automatic return type inference by forcing it to use the
datatype of the shape of the Tensor instead of the fixed Int32.

Including additional tests.
mikeseven pushed a commit to mikeseven/tvm that referenced this pull request Sep 27, 2023
…e#12083)

Fix ArgReduce automatic return type inference by forcing it to use the
datatype of the shape of the Tensor instead of the fixed Int32.

Including additional tests.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Bug] Int64 BroadCast-ArgMax triggers assertion error at graph runtime

4 participants