[REFACTOR] Initialize Unified IR Type Data Structures #4616
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
|
cc @icemelon9 @jroesch @zhiics @wweic @yzhliu @MarisaKirisame @slyubomirsky |
tqchen
force-pushed
the
type
branch
2 times, most recently
from
75a97c1
to
92734a0
Compare
| enum TypeKind : int { | ||
| kType = 0, | ||
| /*! \brief Template variable in shape expression. */ | ||
| kShapeVar = 1, |
There was a problem hiding this comment.
why the separate between kshapevar and kshape?
There was a problem hiding this comment.
I don't know as it was migrated from the original file, we can discuss how to improve in subsequent PRs
This PR moves a few base types from relay to the ir sub-folder. These types will serve as a common type system across the stack. Notably, we want to be able to use the same FuncType for all function signatures. I tried to make a minimum move to bring the necessary dependencies for a FuncType. We can discuss what additional things we want to move as a follow-up. Notably, because the TensorType will have a dependency on low-level Expr, we will need to break the type.h into two files and introduce a tensor_type.h(or leave them in relay for now).
zhiics
approved these changes
Jan 4, 2020
MarisaKirisame
approved these changes
Jan 4, 2020
icemelon
approved these changes
Jan 4, 2020
|
I like the unification of the types for low-level IR and relay. I have one question on the tensor type. Since TE also has Tensor, will it have the TensorType? |
|
To simplify the treatment, most of the low-level expression won't have a tensor type, although by definition it is possible to allow a var to have a tensor type and Pass them to a function, you cannot do anything other than pass it through calls though, because operations like Add, Shuffle only accept normal types. TensorType is also not very well defined because its layout binding are not specified. BufferType might be a better type for such objects in the low-level but we can discuss further. |
|
Thanks @icemelon9 @zhiics @MarisaKirisame |
alexwong
pushed a commit
to alexwong/tvm
that referenced
this pull request
Feb 26, 2020
This PR moves a few base types from relay to the ir sub-folder. These types will serve as a common type system across the stack. Notably, we want to be able to use the same FuncType for all function signatures. I tried to make a minimum move to bring the necessary dependencies for a FuncType. We can discuss what additional things we want to move as a follow-up. Notably, because the TensorType will have a dependency on low-level Expr, we will need to break the type.h into two files and introduce a tensor_type.h(or leave them in relay for now).
alexwong
pushed a commit
to alexwong/tvm
that referenced
this pull request
Feb 28, 2020
This PR moves a few base types from relay to the ir sub-folder. These types will serve as a common type system across the stack. Notably, we want to be able to use the same FuncType for all function signatures. I tried to make a minimum move to bring the necessary dependencies for a FuncType. We can discuss what additional things we want to move as a follow-up. Notably, because the TensorType will have a dependency on low-level Expr, we will need to break the type.h into two files and introduce a tensor_type.h(or leave them in relay for now).
zhiics
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Mar 2, 2020
This PR moves a few base types from relay to the ir sub-folder. These types will serve as a common type system across the stack. Notably, we want to be able to use the same FuncType for all function signatures. I tried to make a minimum move to bring the necessary dependencies for a FuncType. We can discuss what additional things we want to move as a follow-up. Notably, because the TensorType will have a dependency on low-level Expr, we will need to break the type.h into two files and introduce a tensor_type.h(or leave them in relay for now).
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.
Depends on #4615
This PR moves a few base types from relay to the ir sub-folder.
These types will serve as a common type system across the stack.
Notably, we want to be able to use the same FuncType for all function signatures.
I tried to make a minimum move to bring the necessary dependencies for a FuncType.
We can discuss what additional things we want to move as a follow-up.
Because the TensorType will have a dependency on low-level Expr,
we will need to break the type.h into two files and introduce a
tensor_type.h(or leave them in relay for now).