Skip to content

Force double precision in aten.mean.dim#1308

Closed
qedawkins wants to merge 1 commit intollvm:mainfrom
qedawkins:mean_dim_f64
Closed

Force double precision in aten.mean.dim#1308
qedawkins wants to merge 1 commit intollvm:mainfrom
qedawkins:mean_dim_f64

Conversation

@qedawkins
Copy link
Copy Markdown
Collaborator

Accuracy for aten.mean.dim was off for large inputs for the same reason as aten.var.correction (see #1100)

@qedawkins qedawkins force-pushed the mean_dim_f64 branch 3 times, most recently from 831ff2e to a7fddc0 Compare August 29, 2022 20:50
Comment thread lib/Dialect/Torch/Transforms/DecomposeComplexOps.cpp Outdated
@powderluv
Copy link
Copy Markdown
Collaborator

Looks like Refbackend fails to run the test on GHA ? Should we skip it by default ?

@ramiro050
Copy link
Copy Markdown
Collaborator

Is the error not reproducible locally?

qedawkins pushed a commit to nod-ai/torch-mlir that referenced this pull request Oct 3, 2022
qedawkins pushed a commit to nod-ai/torch-mlir that referenced this pull request Oct 3, 2022
The modification to use the builder based interface to generate Krnl loop is completed (llvm#1250, llvm#1283, llvm#1285, llvm#1292, llvm#1293, llvm#1303, llvm#1308, llvm#1314, llvm#1316, llvm#1317, llvm#1326, llvm#1403), and BuildKrnlLoop is no longer needed.

Signed-off-by: Whitney Tsang whitneyt@ca.ibm.com
@qedawkins
Copy link
Copy Markdown
Collaborator Author

Closing as this is fixed by #1605

@qedawkins qedawkins closed this Nov 22, 2022
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.

3 participants