-
Notifications
You must be signed in to change notification settings - Fork 825
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Infer consistent tensor meta #5118
Conversation
oneflow/core/framework/op_interpreter/eager_consistent_op_interpreter.cpp
Outdated
Show resolved
Hide resolved
oneflow/core/framework/op_interpreter/eager_consistent_op_interpreter.cpp
Outdated
Show resolved
Hide resolved
oneflow/core/framework/tensor_meta.h
Outdated
template<> | ||
struct hash<oneflow::one::ConsistentTensorMeta> final { | ||
size_t operator()(const oneflow::one::ConsistentTensorMeta& other) const { | ||
return other.CalcHashValue(); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CalcHashValue
和 hash_value
两种命名有什么区别
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CalcHashValue具有计算过程,hash_value是从对象上拿hash已经存在的值
…infer_consistent_tensor_meta Conflicts: oneflow/core/framework/op_expr.h oneflow/core/framework/tensor.h oneflow/core/framework/tensor_impl.h
…infer_consistent_tensor_meta Conflicts: oneflow/core/framework/tensor_impl.cpp
…flow-Inc/oneflow into infer_consistent_tensor_meta
const UserOpExpr& user_op_expr, const std::vector<BlobDesc>& blob_descs, | ||
std::vector<ParallelDistributionInferHint>* hints) const; | ||
|
||
static Maybe<ConsistentTensorMetaInferArgs> New() { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
把 New 和 Init 合在一起吧,static Maybe<ConsistentTensorMetaInferArgs> New(const TensorTuple& input_tensors, Symbol<PlacementScope> placement_scope, const AttrMap& attrs);
使用ConsistentTensorInferCache来加速推导eager conssitent tensor的元信息。