Put TORCH_LIBRARY in torch/library.h; add custom class API#36742
Closed
ezyang wants to merge 13 commits intogh/ezyang/728/basefrom
Closed
Put TORCH_LIBRARY in torch/library.h; add custom class API#36742ezyang wants to merge 13 commits intogh/ezyang/728/basefrom
ezyang wants to merge 13 commits intogh/ezyang/728/basefrom
Conversation
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
[ghstack-poisoned]
smessmer
approved these changes
Apr 16, 2020
💊 Build failures summary and remediationsAs of commit 41a88fc (more details on the Dr. CI page):
XLA failureJob pytorch_xla_linux_bionic_py3_6_clang9_build is failing. Please create an issue with title prefixed by This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker. This comment has been revised 48 times. |
jamesr66a
reviewed
Apr 16, 2020
| } // namespace jit | ||
|
|
||
| } // namespace torch | ||
| #include <ATen/core/op_registration/custom_class.h> |
Collaborator
There was a problem hiding this comment.
Do we really need to move all this? custom_class.h was meant to be easily accessible and readable for libtorch users and I don't like the indirection this is introducing
Contributor
Author
There was a problem hiding this comment.
I'm not sure. The operant question is whether or not I can include torch/ base directory from ATen/core. I can probably make it work if I push on the build system enough, but my default assumption is it will require a bit of finagling. (Maybe less finagling since custom_class.h already works on mobile, though?)
There was a problem hiding this comment.
Should we ask if the same accessibility concerns apply to the new API as to custom_class.h? Maybe this is a symptom of the new stuff being too buried.
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
[ghstack-poisoned]
ezyang
added a commit
that referenced
this pull request
Apr 16, 2020
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
ghstack-source-id: 3236e79
Pull Request resolved: #36742
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
[ghstack-poisoned]
This was referenced Apr 17, 2020
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom_class_detail.h/custom_class.h were moved to ATen/core, there are
some modifications in custom_class.h: the definition of Library::class_
is in the bottom of that header because I need all of the class_
constructors available, but there is a circular dependency between the
two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
ezyang
added a commit
that referenced
this pull request
Apr 20, 2020
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
ghstack-source-id: 0cfaae8
Pull Request resolved: #36742
Contributor
Author
|
@jamesr66a OK, PR is updated so that |
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
bhosmer
approved these changes
Apr 20, 2020
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
ezyang
added a commit
to pytorch/xla
that referenced
this pull request
Apr 20, 2020
Companion to pytorch/pytorch#36742 Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
```
static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo);
```
you write
```
TORCH_LIBRARY(Namespace, m) {
m.class_<Class>("Class")
.def("foo", foo);
}
```
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Differential Revision: [D21089648](https://our.internmc.facebook.com/intern/diff/D21089648)
[ghstack-poisoned]
ailzhang
pushed a commit
to pytorch/xla
that referenced
this pull request
Apr 21, 2020
* Update include path to torch/library.h Companion to pytorch/pytorch#36742 Signed-off-by: Edward Z. Yang <ezyang@fb.com> * Delete .torch_pin
Contributor
Collaborator
|
Reverting due to master-only breakage of pytorch_linux_xenial_py3_clang5_mobile_code_analysis (see https://app.circleci.com/jobs/github/pytorch/pytorch/5207707) To Run: |
laurentdupin
pushed a commit
to laurentdupin/pytorch
that referenced
this pull request
Apr 24, 2026
…6742) Summary: Pull Request resolved: pytorch#36742 Now, you can define a custom class inside a TORCH_LIBRARY block. It looks very similar to what you did before. Instead of ``` static auto m = torch::class_<Class>("Namespace", "Class").def("foo", foo); ``` you write ``` TORCH_LIBRARY(Namespace, m) { m.class_<Class>("Class") .def("foo", foo); } ``` All the old usages still work, but at some point we should start updating the tutorials when we're ready to go 100% live with the new pybind11 style API. custom class API previously lived in torch/ folder and in torch namespace, so for consistency, the new TORCH_LIBRARY also got moved to torch/library.h The definition of Library::class_ is in the bottom of that header because I need all of the class_ constructors available, but there is a circular dependency between the two headers. Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: D21089648 Test Plan: Imported from OSS Pulled By: ezyang fbshipit-source-id: 8d54329c125242605336c22fa1642aae6940b507
laurentdupin
pushed a commit
to laurentdupin/pytorch
that referenced
this pull request
Apr 24, 2026
…ustom class API" (pytorch#37019) Summary: This reverts commit c062dd8. Original PR: pytorch#36742 Pull Request resolved: pytorch#37019 Differential Revision: D21161361 Pulled By: ezyang fbshipit-source-id: dca4192d3f7be25a34bbe3d57ddce3afc1c2558c
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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.
Stack from ghstack:
Now, you can define a custom class inside a TORCH_LIBRARY block.
It looks very similar to what you did before. Instead of
you write
All the old usages still work, but at some point we should start
updating the tutorials when we're ready to go 100% live with the
new pybind11 style API.
custom class API previously lived in torch/ folder and in torch
namespace, so for consistency, the new TORCH_LIBRARY also got
moved to torch/library.h The definition of Library::class_ is in the
bottom of that header because I need all of the class_ constructors
available, but there is a circular dependency between the two headers.
Signed-off-by: Edward Z. Yang ezyang@fb.com
Differential Revision: D21089648