dmlc--dgl
9699b93136
* Update from master (#4584)
* [Example][Refactor] Refactor graphsage multigpu and full-graph example (#4430)
* Add refactors for multi-gpu and full-graph example
* Fix format
* Update
* Update
* Update
* [Cleanup] Remove async_transferer (#4505)
* Remove async_transferer
* remove test
* Remove AsyncTransferer
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Xin Yao <yaox12@outlook.com>
* [Cleanup] Remove duplicate entries of CUB submodule (issue# 4395) (#4499)
* remove third_part/cub
* remove from third_party
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
* [Bug] Enable turn on/off libxsmm at runtime (#4455)
* enable turn on/off libxsmm at runtime by adding a global config and related API
Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-194.ap-northeast-1.compute.internal>
* [Feature] Unify the cuda stream used in core library (#4480)
* Use an internal cuda stream for CopyDataFromTo
* small fix white space
* Fix to compile
* Make stream optional in copydata for compile
* fix lint issue
* Update cub functions to use internal stream
* Lint check
* Update CopyTo/CopyFrom/CopyFromTo to use internal stream
* Address comments
* Fix backward CUDA stream
* Avoid overloading CopyFromTo()
* Minor comment update
* Overload copydatafromto in cuda device api
Co-authored-by: xiny <xiny@nvidia.com>
* [Feature] Added exclude_self and output_batch to knn graph construction (Issues #4323 #4316) (#4389)
* * Added "exclude_self" and "output_batch" options to knn_graph and segmented_knn_graph
* Updated out-of-date comments on remove_edges and remove_self_loop, since they now preserve batch information
* * Changed defaults on new knn_graph and segmented_knn_graph function parameters, for compatibility; pytorch/test_geometry.py was failing
* * Added test to ensure dgl.remove_self_loop function correctly updates batch information
* * Added new knn_graph and segmented_knn_graph parameters to dgl.nn.KNNGraph and dgl.nn.SegmentedKNNGraph
* * Formatting
* * Oops, I missed the one in segmented_knn_graph when I fixed the similar thing in knn_graph
* * Fixed edge case handling when invalid k specified, since it still needs to be handled consistently for tests to pass
* Fixed context of batch info, since it must match the context of the input position data for remove_self_loop to succeed
* * Fixed batch info resulting from knn_graph when output_batch is true, for case of 3D input tensor, representing multiple segments
* * Added testing of new exclude_self and output_batch parameters on knn_graph and segmented_knn_graph, and their wrappers, KNNGraph and SegmentedKNNGraph, into the test_knn_cuda test
* * Added doc comments for new parameters
* * Added correct handling for uncommon case of k or more coincident points when excluding self edges in knn_graph and segmented_knn_graph
* Added test cases for more than k coincident points
* * Updated doc comments for output_batch parameters for clarity
* * Linter formatting fixes
* * Extracted out common function for test_knn_cpu and test_knn_cuda, to add the new test cases to test_knn_cpu
* * Rewording in doc comments
* * Removed output_batch parameter from knn_graph and segmented_knn_graph, in favour of always setting the batch information, except in knn_graph if x is a 2D tensor
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
* [CI] only known devs are authorized to trigger CI (#4518)
* [CI] only known devs are authorized to trigger CI
* fix if author is null
* add comments
* [Readability] Auto fix setup.py and update-version.py (#4446)
* Auto fix update-version
* Auto fix setup.py
* Auto fix update-version
* Auto fix setup.py
* [Doc] Change random.py to random_partition.py in guide on distributed partition pipeline (#4438)
* Update distributed-preprocessing.rst
* Update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
* fix unpinning when tensoradaptor is not available (#4450)
* [Doc] fix print issue in tutorial (#4459)
* [Example][Refactor] Refactor RGCN example (#4327)
* Refactor full graph entity classification
* Refactor rgcn with sampling
* README update
* Update
* Results update
* Respect default setting of self_loop=false in entity.py
* Update
* Update README
* Update for multi-gpu
* Update
* [doc] fix invalid link in user guide (#4468)
* [Example] directional_GSN for ogbg-molpcba (#4405)
* version-1
* version-2
* version-3
* update examples/README
* Update .gitignore
* update performance in README, delete scripts
* 1st approving review
* 2nd approving review
Co-authored-by: Mufei Li <mufeili1996@gmail.com>
* Clarify the message name, which is 'm'. (#4462)
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
Co-authored-by: Rhett Ying <85214957+Rhett-Ying@users.noreply.github.com>
* [Refactor] Auto fix view.py. (#4461)
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
* [Example] SEAL for OGBL (#4291)
* [Example] SEAL for OGBL
* update index
* update
* fix readme typo
* add seal sampler
* modify set ops
* prefetch
* efficiency test
* update
* optimize
* fix ScatterAdd dtype issue
* update sampler style
* update
Co-authored-by: Quan Gan <coin2028@hotmail.com>
* [CI] use https instead of http (#4488)
* [BugFix] fix crash due to incorrect dtype in dgl.to_block() (#4487)
* [BugFix] fix crash due to incorrect dtype in dgl.to_block()
* fix test failure in TF
* [Feature] Make TensorAdapter Stream Aware (#4472)
* Allocate tensors in DGL's current stream
* make tensoradaptor stream-aware
* replace TAemtpy with cpu allocator
* fix typo
* try fix cpu allocation
* clean header
* redirect AllocDataSpace as well
* resolve comments
* [Build][Doc] Specify the sphinx version (#4465)
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
* reformat
* reformat
* Auto fix update-version
* Auto fix setup.py
* reformat
* reformat
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
Co-authored-by: Rhett Ying <85214957+Rhett-Ying@users.noreply.github.com>
Co-authored-by: Mufei Li <mufeili1996@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Chang Liu <chang.liu@utexas.edu>
Co-authored-by: Zhiteng Li <55398076+ZHITENGLI@users.noreply.github.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: rudongyu <ru_dongyu@outlook.com>
Co-authored-by: Quan Gan <coin2028@hotmail.com>
* Move mock version of dgl_sparse library to DGL main repo (#4524)
* init
* Add api doc for sparse library
* support op btwn matrices with differnt sparsity
* Fixed docstring
* addresses comments
* lint check
* change keyword format to fmt
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
* [DistPart] expose timeout config for process group (#4532)
* [DistPart] expose timeout config for process group
* refine code
* Update tools/distpartitioning/data_proc_pipeline.py
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
* [Feature] Import PyTorch's CUDA stream management (#4503)
* add set_stream
* add .record_stream for NDArray and HeteroGraph
* refactor dgl stream Python APIs
* test record_stream
* add unit test for record stream
* use pytorch's stream
* fix lint
* fix cpu build
* address comments
* address comments
* add record stream tests for dgl.graph
* record frames and update dataloder
* add docstring
* update frame
* add backend check for record_stream
* remove CUDAThreadEntry::stream
* record stream for newly created formats
* fix bug
* fix cpp test
* fix None c_void_p to c_handle
* [examples]educe memory consumption (#4558)
* [examples]educe memory consumption
* reffine help message
* refine
* [Feature][REVIEW] Enable DGL cugaph nightly CI (#4525)
* Added cugraph nightly scripts
* Removed nvcr.io//nvidia/pytorch:22.04-py3 reference
Co-authored-by: Rhett Ying <85214957+Rhett-Ying@users.noreply.github.com>
* Revert "[Feature][REVIEW] Enable DGL cugaph nightly CI (#4525)" (#4563)
This reverts commit ec171c648a.
* [Misc] Add flake8 lint workflow. (#4566)
* Add pyproject.toml for autopep8.
* Add pyproject.toml for autopep8.
* Add flake8 annotation in workflow.
* remove
* add
* clean up
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* [Misc] Try use official pylint workflow. (#4568)
* polish update_version
* update pylint workflow.
* add
* revert.
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* [CI] refine stage logic (#4565)
* [CI] refine stage logic
* refine
* refine
* remove (#4570)
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* Add Pylint workflow for flake8. (#4571)
* remove
* Add pylint.
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* [Misc] Update the python version in Pylint workflow for flake8. (#4572)
* remove
* Add pylint.
* Change the python version for pylint.
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* Update pylint. (#4574)
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* [Misc] Use another workflow. (#4575)
* Update pylint.
* Use another workflow.
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* Update pylint. (#4576)
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* Update pylint.yml
* Update pylint.yml
* Delete pylint.yml
* [Misc]Add pyproject.toml for autopep8 & black. (#4543)
* Add pyproject.toml for autopep8.
* Add pyproject.toml for autopep8.
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
* [Feature] Bump DLPack to v0.7 and decouple DLPack from the core library (#4454)
* rename `DLContext` to `DGLContext`
* rename `kDLGPU` to `kDLCUDA`
* replace DLTensor with DGLArray
* fix linting
* Unify DGLType and DLDataType to DGLDataType
* Fix FFI
* rename DLDeviceType to DGLDeviceType
* decouple dlpack from the core library
* fix bug
* fix lint
* fix merge
* fix build
* address comments
* rename dl_converter to dlpack_convert
* remove redundant comments
Co-authored-by: Chang Liu <chang.liu@utexas.edu>
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Xin Yao <yaox12@outlook.com>
Co-authored-by: Israt Nisa <neesha295@gmail.com>
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
Co-authored-by: peizhou001 <110809584+peizhou001@users.noreply.github.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-194.ap-northeast-1.compute.internal>
Co-authored-by: ndickson-nvidia <99772994+ndickson-nvidia@users.noreply.github.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: Rhett Ying <85214957+Rhett-Ying@users.noreply.github.com>
Co-authored-by: Hongzhi (Steve), Chen <chenhongzhi.nkcs@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
Co-authored-by: Zhiteng Li <55398076+ZHITENGLI@users.noreply.github.com>
Co-authored-by: rudongyu <ru_dongyu@outlook.com>
Co-authored-by: Quan Gan <coin2028@hotmail.com>
Co-authored-by: Vibhu Jawa <vibhujawa@gmail.com>
* [Deprecation] Dataset Attributes (#4546)
* Update
* CI
* CI
* Update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
* [Example] Bug Fix (#4665)
* Update
* CI
* CI
* Update
* Update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
* Update
* Update (#4724)
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
* change DGLHeteroGraph to DGLGraph in DOC
* revert c change
Co-authored-by: Mufei Li <mufeili1996@gmail.com>
Co-authored-by: Chang Liu <chang.liu@utexas.edu>
Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Xin Yao <yaox12@outlook.com>
Co-authored-by: Israt Nisa <neesha295@gmail.com>
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-194.ap-northeast-1.compute.internal>
Co-authored-by: ndickson-nvidia <99772994+ndickson-nvidia@users.noreply.github.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: Rhett Ying <85214957+Rhett-Ying@users.noreply.github.com>
Co-authored-by: Hongzhi (Steve), Chen <chenhongzhi.nkcs@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-26.ap-northeast-1.compute.internal>
Co-authored-by: Zhiteng Li <55398076+ZHITENGLI@users.noreply.github.com>
Co-authored-by: rudongyu <ru_dongyu@outlook.com>
Co-authored-by: Quan Gan <coin2028@hotmail.com>
Co-authored-by: Vibhu Jawa <vibhujawa@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-16-19.ap-northeast-1.compute.internal>
274 行
13 KiB
ReStructuredText
274 行
13 KiB
ReStructuredText
.. _guide_ko-training-edge-classification:
|
|
|
|
5.2 에지 분류 및 리그레션(Regression)
|
|
--------------------------------
|
|
|
|
:ref:`(English Version) <guide-training-edge-classification>`
|
|
|
|
때론 그래프의 에지들의 속성을 예측을 원하는 경우가 있다. 이를 위해서 *에지 분류/리그레션* 모델을 만들고자 한다.
|
|
|
|
우선, 예제로 사용할 에지 예측을 위한 임의의 그래프를 만든다.
|
|
|
|
.. code:: python
|
|
|
|
src = np.random.randint(0, 100, 500)
|
|
dst = np.random.randint(0, 100, 500)
|
|
# make it symmetric
|
|
edge_pred_graph = dgl.graph((np.concatenate([src, dst]), np.concatenate([dst, src])))
|
|
# synthetic node and edge features, as well as edge labels
|
|
edge_pred_graph.ndata['feature'] = torch.randn(100, 10)
|
|
edge_pred_graph.edata['feature'] = torch.randn(1000, 10)
|
|
edge_pred_graph.edata['label'] = torch.randn(1000)
|
|
# synthetic train-validation-test splits
|
|
edge_pred_graph.edata['train_mask'] = torch.zeros(1000, dtype=torch.bool).bernoulli(0.6)
|
|
|
|
개요
|
|
~~~~~~~~~
|
|
|
|
앞 절에서 우리는 멀티 레이어 GNN을 사용해서 노드 분류하는 방법을 알아봤다. 임의의 노드에 대한 hidden representation을 계산하기 위해서 같은 기법을 적용한다. 그러면 에지들에 대한 예측은 그것들의 부속 노드들의 representation들로 부터 도출할 수 있다.
|
|
|
|
에지에 대한 예측을 계산하는 가장 일반적인 방법은 그 에지의 부속 노드들의 representation들과 부수적으로 그 에지에 대한 피쳐들의 parameterized 함수로 표현하는 것이다.
|
|
|
|
노드 분류 모델과 구현상의 차이점
|
|
~~~~~~~~~~~~~~~~~~~~~~~~
|
|
|
|
이전 절에서 만든 모델을 사용해서 노드 representation을 계산한다고 가정하면, :meth:`~dgl.DGLGraph.apply_edges` 메소드로 에지 예측을 계산하는 컴포넌트만 작성하면 된다.
|
|
|
|
예를 들어, 에지 리그레션을 위해서 각 에지에 대한 점수를 계산하고자 한다면, 아래 코드와 같이 각 에지에 대한 부속 노드의 representation들의 dot product를 계산하면 된다.
|
|
|
|
.. code:: python
|
|
|
|
import dgl.function as fn
|
|
class DotProductPredictor(nn.Module):
|
|
def forward(self, graph, h):
|
|
# h contains the node representations computed from the GNN defined
|
|
# in the node classification section (Section 5.1).
|
|
with graph.local_scope():
|
|
graph.ndata['h'] = h
|
|
graph.apply_edges(fn.u_dot_v('h', 'h', 'score'))
|
|
return graph.edata['score']
|
|
|
|
또한 MLP를 사용해서 각 에지에 대한 벡터 값을 예측하는 예측하는 함수를 작성할 수도 있다. 이 벡터 값은 미래의 다운스트림 테스크들에 사용될 수 있다. 즉, 범주형 분류의 logit으로 사용.
|
|
|
|
.. code:: python
|
|
|
|
class MLPPredictor(nn.Module):
|
|
def __init__(self, in_features, out_classes):
|
|
super().__init__()
|
|
self.W = nn.Linear(in_features * 2, out_classes)
|
|
|
|
def apply_edges(self, edges):
|
|
h_u = edges.src['h']
|
|
h_v = edges.dst['h']
|
|
score = self.W(torch.cat([h_u, h_v], 1))
|
|
return {'score': score}
|
|
|
|
def forward(self, graph, h):
|
|
# h contains the node representations computed from the GNN defined
|
|
# in the node classification section (Section 5.1).
|
|
with graph.local_scope():
|
|
graph.ndata['h'] = h
|
|
graph.apply_edges(self.apply_edges)
|
|
return graph.edata['score']
|
|
|
|
학습 룹(loop)
|
|
~~~~~~~~~~~
|
|
|
|
노드 representation 계산 모델과 에지 예측 모델을 만들었다면, 모든 에지들에 대한 예측값을 계산하는 전체 그래프를 이용한 학습 룹을 작성할 수 있다.
|
|
|
|
노드 representation 계산 모델로 ``SAGE`` 를, 에지 예측 모델로 ``DotPredictor`` 을 사용한다.
|
|
|
|
.. code:: python
|
|
|
|
class Model(nn.Module):
|
|
def __init__(self, in_features, hidden_features, out_features):
|
|
super().__init__()
|
|
self.sage = SAGE(in_features, hidden_features, out_features)
|
|
self.pred = DotProductPredictor()
|
|
def forward(self, g, x):
|
|
h = self.sage(g, x)
|
|
return self.pred(g, h)
|
|
|
|
이 예제에서 학습/검증/테스트 에지 셋이 에지의 이진 마스크로 구분된다고 가정한다. 또한 early stopping이나 모델 저장은 포함하지 않는다.
|
|
|
|
.. code:: python
|
|
|
|
node_features = edge_pred_graph.ndata['feature']
|
|
edge_label = edge_pred_graph.edata['label']
|
|
train_mask = edge_pred_graph.edata['train_mask']
|
|
model = Model(10, 20, 5)
|
|
opt = torch.optim.Adam(model.parameters())
|
|
for epoch in range(10):
|
|
pred = model(edge_pred_graph, node_features)
|
|
loss = ((pred[train_mask] - edge_label[train_mask]) ** 2).mean()
|
|
opt.zero_grad()
|
|
loss.backward()
|
|
opt.step()
|
|
print(loss.item())
|
|
|
|
.. _guide_ko-training-edge-classification-heterogeneous-graph:
|
|
|
|
Heterogeneous 그래프
|
|
~~~~~~~~~~~~~~~~~~
|
|
|
|
Heterogeneous 그래프들에 대한 에지 분류는 homogeneous 그래프와 크게 다르지 않다. 하나의 에지 타입에 대해서 에지 분류를 수행하자 한다면, 모든 노드 티압에 대한 노드 representation을 구하고, :meth:`~dgl.DGLGraph.apply_edges` 메소드를 사용해서 에지 타입을 예측하면 된다.
|
|
|
|
예를 들면, heterogeneous 그래프의 하나의 에지 타입에 대한 동작하는 ``DotProductPredictor`` 를 작성하고자 한다면, ``apply_edges`` 메소드에 해당 에지 타입을 명시하기만 하면 된다.
|
|
|
|
.. code:: python
|
|
|
|
class HeteroDotProductPredictor(nn.Module):
|
|
def forward(self, graph, h, etype):
|
|
# h contains the node representations for each edge type computed from
|
|
# the GNN for heterogeneous graphs defined in the node classification
|
|
# section (Section 5.1).
|
|
with graph.local_scope():
|
|
graph.ndata['h'] = h # assigns 'h' of all node types in one shot
|
|
graph.apply_edges(fn.u_dot_v('h', 'h', 'score'), etype=etype)
|
|
return graph.edges[etype].data['score']
|
|
|
|
비슷하게 ``HeteroMLPPredictor`` 를 작성할 수 있다.
|
|
|
|
.. code:: python
|
|
|
|
class HeteroMLPPredictor(nn.Module):
|
|
def __init__(self, in_features, out_classes):
|
|
super().__init__()
|
|
self.W = nn.Linear(in_features * 2, out_classes)
|
|
|
|
def apply_edges(self, edges):
|
|
h_u = edges.src['h']
|
|
h_v = edges.dst['h']
|
|
score = self.W(torch.cat([h_u, h_v], 1))
|
|
return {'score': score}
|
|
|
|
def forward(self, graph, h, etype):
|
|
# h contains the node representations for each edge type computed from
|
|
# the GNN for heterogeneous graphs defined in the node classification
|
|
# section (Section 5.1).
|
|
with graph.local_scope():
|
|
graph.ndata['h'] = h # assigns 'h' of all node types in one shot
|
|
graph.apply_edges(self.apply_edges, etype=etype)
|
|
return graph.edges[etype].data['score']
|
|
|
|
특정 타입의 에지에 대해서, 각 에지의 점수를 예측하는 end-to-end 모델을 다음과 같다:
|
|
|
|
.. code:: python
|
|
|
|
class Model(nn.Module):
|
|
def __init__(self, in_features, hidden_features, out_features, rel_names):
|
|
super().__init__()
|
|
self.sage = RGCN(in_features, hidden_features, out_features, rel_names)
|
|
self.pred = HeteroDotProductPredictor()
|
|
def forward(self, g, x, etype):
|
|
h = self.sage(g, x)
|
|
return self.pred(g, h, etype)
|
|
|
|
모델을 사용하는 방법은 노드 타입과 피쳐들에 대한 사전을 모델에 간단하게 입력하면 된다.
|
|
|
|
.. code:: python
|
|
|
|
model = Model(10, 20, 5, hetero_graph.etypes)
|
|
user_feats = hetero_graph.nodes['user'].data['feature']
|
|
item_feats = hetero_graph.nodes['item'].data['feature']
|
|
label = hetero_graph.edges['click'].data['label']
|
|
train_mask = hetero_graph.edges['click'].data['train_mask']
|
|
node_features = {'user': user_feats, 'item': item_feats}
|
|
|
|
학습 룹은 homogeneous 그래프의 것과 거의 유사하다. 예를 들어, 에지 타입 ``click`` 에 대한 에지 레이블을 예측하는 것은 다음과 같이 간단히 구현된다.
|
|
|
|
.. code:: python
|
|
|
|
opt = torch.optim.Adam(model.parameters())
|
|
for epoch in range(10):
|
|
pred = model(hetero_graph, node_features, 'click')
|
|
loss = ((pred[train_mask] - label[train_mask]) ** 2).mean()
|
|
opt.zero_grad()
|
|
loss.backward()
|
|
opt.step()
|
|
print(loss.item())
|
|
|
|
|
|
Heterogeneous 그래프의 에지들에 대한 에지 타입 예측하기
|
|
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
|
|
|
주어진 에지의 타입을 예측하는 일도 종종 하게된다.
|
|
|
|
:ref:`heterogeneous 그래프 예제 <guide_ko-training-heterogeneous-graph-example>` 에서는 user와 item을 연결하는 에지가 주어졌을 때, user가 ``click`` 을 선택할지, ``dislike`` 를 선택할지를 예측하고 있다.
|
|
|
|
이는 추천에서 흔히 쓰이는 평가 예측의 간략한 버전이다.
|
|
|
|
노드 representation을 얻기 위해서 heterogeneous graph convolution 네트워크를 사용할 수 있다. 이를 위해서 :ref:`이전에 정의한 RGCN <guide_ko-training-rgcn-node-classification>` 를 사용하는 것도 가능하다.
|
|
|
|
에지 타입을 예측하기 위해서 ``HeteroDotProductPredictor`` 의 용도를 간단히 변경해서 예측할 모든 에지 타입을 “병합“하고 모든 에지들의 각 타입에 대한 점수를 내보내는 하나의 에지 타입만 있는 다른 그래프를 취하게하면 된다.
|
|
|
|
이 예제에 적용해보면, ``user`` 와 ``item`` 두 노트 타입을 갖으며 ``user`` 와 ``item`` 에 대한 ``click`` 이나 ``dislike`` 같은 모든 에지 타입을 병합하는 단일 에지 타입을 갖는 그래프가 필요하다. 다음 문장으로 간단하게 생성할 수 있다.
|
|
|
|
.. code:: python
|
|
|
|
dec_graph = hetero_graph['user', :, 'item']
|
|
|
|
이 함수는 ``user`` 와 ``item`` 을 노드 타입으로 갖고, 두 노드 타입을 연결하고 있는 모든 에지 타입(예, ``click`` 와 ``dislike`` )을 합친 단일 에지 타입을 갖는 heterogeneous 그래프를 리턴한다.
|
|
|
|
위 코드는 원래의 에지 타입을 ``dgl.ETYPE`` 이라는 이름의 피처로 리턴하기 때문에, 이를 레이블로 사용할 수 있다.
|
|
|
|
.. code:: python
|
|
|
|
edge_label = dec_graph.edata[dgl.ETYPE]
|
|
|
|
에지 타입 예측 모듈의 입력으로 위 그래프를 사용해서 예측 모델을 다음과 같이 작성한다.
|
|
|
|
.. code:: python
|
|
|
|
class HeteroMLPPredictor(nn.Module):
|
|
def __init__(self, in_dims, n_classes):
|
|
super().__init__()
|
|
self.W = nn.Linear(in_dims * 2, n_classes)
|
|
|
|
def apply_edges(self, edges):
|
|
x = torch.cat([edges.src['h'], edges.dst['h']], 1)
|
|
y = self.W(x)
|
|
return {'score': y}
|
|
|
|
def forward(self, graph, h):
|
|
# h contains the node representations for each edge type computed from
|
|
# the GNN for heterogeneous graphs defined in the node classification
|
|
# section (Section 5.1).
|
|
with graph.local_scope():
|
|
graph.ndata['h'] = h # assigns 'h' of all node types in one shot
|
|
graph.apply_edges(self.apply_edges)
|
|
return graph.edata['score']
|
|
|
|
노드 representation 모듈과 에지 타입 예측 모듈을 합친 모델은 다음과 같다.
|
|
|
|
.. code:: python
|
|
|
|
class Model(nn.Module):
|
|
def __init__(self, in_features, hidden_features, out_features, rel_names):
|
|
super().__init__()
|
|
self.sage = RGCN(in_features, hidden_features, out_features, rel_names)
|
|
self.pred = HeteroMLPPredictor(out_features, len(rel_names))
|
|
def forward(self, g, x, dec_graph):
|
|
h = self.sage(g, x)
|
|
return self.pred(dec_graph, h)
|
|
|
|
학습 룹은 아래와 같이 간단하다.
|
|
|
|
.. code:: python
|
|
|
|
model = Model(10, 20, 5, hetero_graph.etypes)
|
|
user_feats = hetero_graph.nodes['user'].data['feature']
|
|
item_feats = hetero_graph.nodes['item'].data['feature']
|
|
node_features = {'user': user_feats, 'item': item_feats}
|
|
|
|
opt = torch.optim.Adam(model.parameters())
|
|
for epoch in range(10):
|
|
logits = model(hetero_graph, node_features, dec_graph)
|
|
loss = F.cross_entropy(logits, edge_label)
|
|
opt.zero_grad()
|
|
loss.backward()
|
|
opt.step()
|
|
print(loss.item())
|
|
|
|
DGL은 heterogeneous 그래프의 에지들에 대한 타입을 예측하는 문제인 평가 예측 예제로 `Graph Convolutional Matrix Completion <https://github.com/dmlc/dgl/tree/master/examples/pytorch/gcmc>`__ 를 제공한다. `모델 구현 파일 <https://github.com/dmlc/dgl/tree/master/examples/pytorch/gcmc>`__ 에 있는 노드 representation 모듈은 ``GCMCLayer`` 라고 불린다. 이 둘은 여기서 설명하기에는 너무 복잡하니 자세한 설명은 생략한다.
|