[LANG] Generalize compute to tensor region #1476
Merged
Conversation
This file contains 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
|
@ZihengJiang can you first open an RFC issue discussing the proposal and designs, examples, then link to the PR? |
|
We have some operators based on tensor op, also we've done lots of testing for this api, could this be merged into master ASAP? @tqchen @ZihengJiang |
|
@ZihengJiang could you please fix cpp lint first? |
ZihengJiang
force-pushed
the
tensor_op
branch
2 times, most recently
from
27dbfbe to
f892571
Compare
xqdan
reviewed
Aug 10, 2018
| # test_tensor_reduce() | ||
| test_tensor_tensor_op() | ||
| # test_tensor_scan() | ||
| # test_scan_multi_out() |
There was a problem hiding this comment.
you need to enable these cases
zhiics
reviewed
Aug 10, 2018
src/op/tensor_op.cc
Outdated
|
|
||
| Operation TensorOpNode::make(std::string name, | ||
| std::string tag, | ||
| Array<IterVar> axis, |
ZihengJiang
force-pushed
the
tensor_op
branch
from
35bdd6e to
1dfa7b4
Compare
ZihengJiang
added
status: need review
and removed
status: need RFC
ZihengJiang
force-pushed
the
tensor_op
branch
from
f176f43 to
cf0134d
Compare
|
@were @yzhliu @merrymercy can you help review this PR ? |
xqdan
reviewed
Sep 29, 2018
tqchen
requested changes
Oct 1, 2018
include/tvm/operation.h
Outdated
| @@ -182,6 +182,80 @@ class PlaceholderOpNode : public OperationNode { | |||
| TVM_DECLARE_NODE_TYPE_INFO(PlaceholderOpNode, OperationNode); | |||
| }; | |||
|
|
|||
| class TensorComputeOpNode : public OperationNode { | |||
| public: | |||
| Array<IterVar> axis; | |||
tqchen
requested changes
Oct 4, 2018
include/tvm/operation.h
Outdated
| v->Visit("reduce_axis", &reduce_axis); | ||
| v->Visit("sch_ndim", &sch_ndim); |
include/tvm/operation.h
Outdated
| @@ -186,12 +186,14 @@ class TensorComputeOpNode : public OperationNode { | |||
| public: | |||
| Array<IterVar> axis; | |||
| @@ -154,6 +155,7 @@ def intrin_func(ins, outs): | |||
|
|
|||
| s = tvm.create_schedule(C.op) | |||
| stmt = tvm.lower(s, [A, B, C], simple_mode=True) | |||
| print(stmt) | |||
tqchen
added
status: need update
tqchen
approved these changes
Oct 6, 2018
|
Thanks @ZihengJiang @xqdan @yzhliu @zhiics this is now merged |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Thanks for contributing to TVM! Please refer to guideline https://docs.tvm.ai/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from others in the community.
Another way for tensor intrinsic