Skip to content

Minor refactor for temporal sampling #257

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.

Sign up for GitHub

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

Merged
merged 18 commits into from
Jul 16, 2022
Merged

Minor refactor for temporal sampling #257

Changes from all commits
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
52f9e9d
disable undirected for temporal sampling
RexYing Jul 1, 2022
e66b65c
disjoint sampling for temporal
RexYing Jul 1, 2022
748f816
fix repeated node index
RexYing Jul 1, 2022
628f0f6
compile fix
RexYing Jul 1, 2022
7fce422
Update csrc/cpu/neighbor_sample_cpu.cpp
RexYing Jul 15, 2022
e0076a6
Update csrc/cpu/neighbor_sample_cpu.cpp
RexYing Jul 15, 2022
8f203eb
Update csrc/cpu/neighbor_sample_cpu.cpp
RexYing Jul 15, 2022
c74f15b
Merge branch 'master' into temporal_sample_disable_undirected
RexYing Jul 15, 2022
e98496f
comments on directed to be true
RexYing Jul 15, 2022
28a577e
add directed in API
RexYing Jul 16, 2022
61e3167
comments
RexYing Jul 16, 2022
63b16e0
minor function signature fix
RexYing Jul 16, 2022
efb9776
Update csrc/cpu/neighbor_sample_cpu.cpp
rusty1s Jul 16, 2022
9595ea3
Update csrc/neighbor_sample.cpp
rusty1s Jul 16, 2022
7059aa7
Merge branch 'temporal_sample_disable_undirected' of github.com:RexYi…
RexYing Jul 16, 2022
04d1eb5
Merge branch 'master' into temporal_sample_disable_undirected
RexYing Jul 16, 2022
0769573
minor refactor
RexYing Jul 16, 2022
2584d98
minor refactor
RexYing Jul 16, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions csrc/cpu/neighbor_sample_cpu.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -242,10 +242,11 @@ hetero_sample(const vector<node_t> &node_types,
// note that the sampling always needs to have directed=True
// for temporal case
// to_local_src_node is not used for temporal / directed case
const int64_t sample_idx = src_samples.size();
src_samples.push_back(v);
src_root_time.push_back(dst_time);
cols.push_back(i);
rows.push_back(src_samples.size() - 1);
rows.push_back(sample_idx);
edges.push_back(offset);
} else {
const auto res = to_local_src_node.insert({v, src_samples.size()});
Expand All @@ -271,10 +272,11 @@ hetero_sample(const vector<node_t> &node_types,
// force disjoint of computation tree
// note that the sampling always needs to have directed=True
// for temporal case
const int64_t sample_idx = src_samples.size();
src_samples.push_back(v);
src_root_time.push_back(dst_time);
cols.push_back(i);
rows.push_back(src_samples.size() - 1);
rows.push_back(sample_idx);
edges.push_back(offset);
} else {
const auto res = to_local_src_node.insert({v, src_samples.size()});
Expand Down Expand Up @@ -305,10 +307,11 @@ hetero_sample(const vector<node_t> &node_types,
// force disjoint of computation tree
// note that the sampling always needs to have directed=True
// for temporal case
const int64_t sample_idx = src_samples.size();
src_samples.push_back(v);
src_root_time.push_back(dst_time);
cols.push_back(i);
rows.push_back(src_samples.size() - 1);
rows.push_back(sample_idx);
edges.push_back(offset);
} else {
const auto res = to_local_src_node.insert({v, src_samples.size()});
Expand Down Expand Up @@ -431,7 +434,7 @@ hetero_temporal_neighbor_sample_cpu(
const c10::Dict<rel_t, vector<int64_t>> &num_neighbors_dict,
const c10::Dict<node_t, torch::Tensor> &node_time_dict,
const int64_t num_hops, const bool replace, const bool directed) {
AT_ASSERTM(directed, "Temporal sampling requires 'directed' sampling")
AT_ASSERTM(directed, "Temporal sampling requires 'directed' sampling");
if (replace) {
// We assume that directed = True for temporal sampling
// The current implementation uses disjoint computation trees
Expand Down