cann: update cross_entropy_loss op support #16886
Conversation
github-actions
bot
added
ggml
|
Could you paste your accuracy test results? |
noemotiovon
left a comment
noemotiovon
left a comment
There was a problem hiding this comment.
There’s a minor issue here; I’ll review it more thoroughly later.
| ggml_cann_group_norm(ctx, dst); | ||
| break; | ||
| case GGML_OP_L2_NORM: | ||
| ggml_cann_l2_norm(ctx, dst); |
There was a problem hiding this comment.
You’ve already added this in another PR. I’ll take a closer look at your PR after that one gets merged.
ggml/src/ggml-cann/ggml-cann.cpp
Outdated
| return true; | ||
| } | ||
| default: | ||
| std::cout << "Unsupported op: " << ggml_op_name(op->op) << std::endl; |
There was a problem hiding this comment.
I think this print statement was left in by mistake.
There was a problem hiding this comment.
ok, I will delete it.
|
You need rebase to pass the failing test. |
…operator that already exists in another pull request.
TecJesh
force-pushed
the
cross_entropy_loss
branch
from
5a6bf29 to
1d3d26a
Compare
|
Backend 1/2: CANN0 new_pool_for_device: device 0 use vmm pool |
3 participants
The cann backend currently does not support computing cross_entropy_loss; add support for the cross_entropy_loss operator to the cann computation backend.