DistABLPLoader: vectorized label remap + optional AnchorLabels edge-list output by kmontemayor2-sc · Pull Request #681 · Snapchat/GiGL

kmontemayor2-sc · 2026-06-25T14:45:00Z

Summary

Reworks how DistABLPLoader remaps Anchor-Based Link Prediction (ABLP) labels
from global node ids to subgraph-local indices, and adds an optional edge-list
output format for those labels.

At prod scale, we see 60% speedup for ABLP next time for colcated and 30% for graph store (e.g. 40 -> 15 and 60 -> 40).

Internally, we see small but

Vectorized label remap. Label remapping now runs as a single
sorted-membership join (searchsorted over the node map) instead of a
per-anchor Python loop. The result is loss-equivalent to the previous output
(same (anchor, label) pairs; within-anchor order is unspecified, and the
ABLP contrastive loss is order-invariant). We observe meaningful sampling/
collation performance improvements at production scale.
AnchorLabels edge-list container + use_list_output flag.
DistABLPLoader(..., use_list_output=True) returns labels as an
AnchorLabels edge list (two co-indexed [E] tensors,
anchor_index / label_index) that the loss can index directly with no
padding or per-anchor Python loop. use_list_output=False (default) preserves
the existing ragged dict[int, torch.Tensor] output, so this is backward
compatible. AnchorLabels.to_dict() recovers the dict form.
Public API. AnchorLabels is exported from gigl.distributed.
Examples. The link-prediction training examples (homogeneous +
heterogeneous, colocated + graph-store) are updated to consume the
AnchorLabels edge-list output.

Tests

Equivalence of the vectorized remap output against constructed expected values,
covering empty / fully-padded / duplicate-label / multi-edge-type cases.
use_list_output=True vs =False produce equivalent labels.
CUDA device-placement regression test for the remap kernel.
Guard: a non-unique node→global map raises ValueError.

Notes

Labels are loss-equivalent to the prior output, not order-identical.
Documentation-only follow-ups: worked examples for both output formats in the
DistABLPLoader docstring and AnchorLabels shape docs.

…racle Factor the inline per-anchor label-remap loop in DistABLPLoader._set_labels into a module-level function _loop_set_labels. The new function is a behavior-preserving extraction: _set_labels delegates to it, producing identical output. _loop_set_labels will serve as the equivalence oracle for the vectorized kernel added in the next task. Also imports PADDING_NODE from gigl.utils.data_splitters (used by the vectorized kernel in the next task) and adds the contract test file tests/unit/distributed/vectorized_set_labels_test.py. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…oop oracle Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…ls kernel Add frozen dataclass AnchorLabels (anchor_index, label_index, num_anchors) with to_dict() bridge; thin wrapper _remap_one_label_tensor_edge_list over the shared _membership_remap; and edge_list_set_labels driver. Proves to_dict() reproduces _loop_set_labels bit-for-bit via parametrized equivalence tests (11 new tests, all classes pass). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…eous training Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…eneous training Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…test Line-wrapping only (ruff format) for the edge-list label reads in the four link-prediction training examples and the loader equivalence test. No logic change. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…ctors **Behavioral change:** Drop the composite-key stable argsort from `_membership_remap` (the `composite_key = anchor_kept * (num_nodes + 1) + local_idx` / `torch.argsort(composite_key, stable=True)` block). The pair stream is now emitted in column-visit (row-major masked flatten) order rather than ascending-local-index (torch.nonzero) order. **Why this is safe:** `RetrievalLoss` (`gigl/nn/loss.py`) is `CrossEntropyLoss(reduction="sum")` over a diagonal-targeted score matrix that masks collisions by id VALUE, not position. It is invariant to any joint permutation of `(anchor_index, label_index)` pairs. The only constraint is co-indexing (pair k stays intact) and per-anchor grouping, both of which are preserved. Within-anchor label order was an implementation artifact of the loop oracle -- never a loss requirement. **Why the dict path is unaffected:** `anchor_of_entry` is built as `arange(N).repeat_interleave(M)` (row-major), so the masked flatten is already non-decreasing in anchor_index. `bincount`/`split` requires only contiguous grouping by anchor, which row-major flatten guarantees without any argsort. **Readability refactors in `_membership_remap`:** - Rename terse tensors: `valid` -> `is_present`, `found` -> `is_exact_match`, `positions` -> `sorted_positions`, `local_idx` -> `local_index`, `anchor_kept` -> `anchor_of_matched` - Add numbered step comments explaining the searchsorted membership lookup **Readability refactors in outer kernels:** - Add `_remap_group` helper to collapse the ~40-line duplicated positive/ negative per-edge-type loops in both `vectorized_set_labels` and `edge_list_set_labels` (uses TypeVar for generic return type) - `_sorted_for` closure pattern preserved, inlined per-kernel (memoizes torch.sort across pos/neg edge types of the same node type) **Other improvements:** - Add doctest to `AnchorLabels` showing a 2-anchor case + `to_dict()` round-trip - Update `AnchorLabels` docstring: document column-visit order and loss-permutation-invariance rationale - Update all docstrings to drop "bit-for-bit"/"torch.nonzero order" language **Test contract relaxation (tests remain non-vacuous):** - `vectorized_set_labels_test.py` / `edge_list_set_labels_test.py`: `_assert_label_dicts_equal` -> `_assert_label_dicts_set_equal` (uses `sorted()` per anchor; still catches membership errors + multiplicity) - `test_unsorted_node_map_exact_order` -> `test_unsorted_node_map_correct_membership`: asserts SET {0, 2} instead of sequence [0, 2]; docstring explains column vs ascending-local order difference - `RemapOneEdgeListTest.test_unsorted_node_map_nontrivial_sort_perm` -> `test_unsorted_node_map_correct_membership`: pins column order [2, 0] and explains it is SET-equal to [0, 2]; verifies sort_perm mapping is correct - `dist_ablp_neighborloader_test.py`: `_ordered_global_pairs` -> `_global_pair_set` (sorts within anchor); `_collect_homogeneous_labels` docstring updated; the in-process exact-tensor assertion (`label_index == cat(dict values)`) is preserved -- both paths draw from the same `_membership_remap` pair stream so they remain identical - CUDA device test docstring: remove "stable-argsort tie-break" reference; keep the duplicate [15,15] row (still tests duplicate handling + device placement) Device fix preserved: `anchor_of_entry` is still built on `label_tensor.device`. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

…er-identical After dropping the order-reproduction argsort, per-anchor label order is column-visit order, not the loop's nonzero order. The (query, label) pairs are unchanged and the contrastive loss is order-invariant, so the read is equivalent for training; the comments now say so instead of implying byte-identical order. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

GiGL main resolved ABLP labels with a per-anchor Python loop. This replaces it with one vectorized kernel and a single output type, plus a thin dict view for backward compatibility. - Resolve labels with a sorted-membership join (_membership_remap: sort the node map once, searchsorted the label ids, keep exact matches) instead of the O(N_anchors * M * N_nodes) per-anchor loop. Delete _loop_set_labels (no production callers) and the redundant dict-producing kernel. - Collapse the remap to two functions, _membership_remap and edge_list_set_labels; the dict path is just AnchorLabels.to_dict(), selected by use_list_output on DistABLPLoader (default False keeps the ragged dict). Drop the _remap_group callback indirection, the _LabelT TypeVar, and the vestigial supervision_edge_types parity param. - AnchorLabels stores labels as two parallel (anchor_index, label_index) tensors so the loss can index them directly; within-anchor order is unspecified because the ABLP contrastive loss is order-invariant over the pairs. - Make the __debug__ unique-node-map check a cheap adjacent-difference test on the already-sorted map (it ran torch.unique every batch, and GiGL is not launched with -O). - Tests assert against constructed expected values and per-anchor label SETS, not within-anchor order. Examples consume the edge-list directly. Behavior-preserving: per-anchor label sets are unchanged. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…ions Comments and docstrings only -- no behavior change (verified: the AST with docstrings stripped is identical to the prior commit for every touched file). - Docstrings lead with why over what; the membership-remap algorithm is shown as a small worked example (node map + [N_anchors, M] padded label tensor -> pairs) with labeled steps instead of dense prose. - Inline comments reference those docstring steps rather than floating free. - Tensor dimensions annotated throughout (N_anchors / M / N_nodes / K / E), including the K (non-padding candidates) vs E (matched pairs) distinction. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

…uniqueness guard - Add a use_list_output=True worked example to DistABLPLoader.__init__ that mirrors the use_list_output=False example (same graph, AnchorLabels values). - Document AnchorLabels.to_dict()'s non-decreasing anchor_index precondition. - Promote the _membership_remap duplicate-node-map guard from a __debug__ assert to an always-on ValueError so it still fires under `python -O`; update the test to expect ValueError. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Comments should describe the code as it is, not as a delta from a prior or never-committed version. Reword/remove comments that only parse if the reader knows the code's development history: - Drop the "Always-on (not __debug__) ... python -O" framing on the node-map uniqueness guard; keep the precondition + empty-slice safety rationale. - Fix a stale _membership_remap docstring line still describing a __debug__ assertion (it is now an always-on ValueError). - Remove "with no argsort" / "no argsort needed" (presupposed the dropped order-reproduction argsort); state the grouped-by-anchor property directly. - Reword "a sorted-membership join rather than a per-anchor scan ... trades a broadcast-compare for a search" to describe the algorithm and its complexity as-is. - Drop "rather than maintaining a second kernel" in _set_labels. - Same scrub in edge_list_set_labels_test.py (module docstring + guard test). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

mkolodner-sc

Thanks a lot for looking into this Kyle! This is a super exciting change, and looking forward to realizing the throughput and general complexity improvements from this work.

Did a first pass here through the production logic and left some comments.

mkolodner-sc · 2026-07-27T10:36:21Z

+        num_anchors (int): Total number of anchors ``N_anchors`` (rows of the
+            source padded label tensor), including anchors with no labels.


Do we need the num_anchors data_class? This should already be available in Data/HeteroData object right

mkolodner-sc · 2026-07-27T10:41:05Z

+        anchor_index (torch.Tensor): ``[E]`` long tensor of local anchor rows.
+        label_index (torch.Tensor): ``[E]`` long tensor of local label node ids.


I'm wondering if we can represent this instead as a [2, E] edge tensor, since that is essentially what it seems to be used for here. This would make it so we'd only need one .to(device) call. If we can make this just one tensor instead of two and we can remove the num_anchors field as well (which already exists in the HeteroData), I'm wondering if we can remove the need for a dataclass here all together.

The benefit is that the resulting labeled object would just be an edge tensor and would feel more PyG-esque and natural to use, rather than a GiGL-specific dataclass for labeled objects. A [2, E] (labeled) edge index tensor also feels similar to what we use for other objects such as the message passing edge index tensor or PPR edges, which we don't have custom dataclasses for.

mkolodner-sc · 2026-07-27T10:50:09Z

 logger = Logger()


+@dataclass(frozen=True)


Could we move the newly introduced remapping logic to a separate utility file, maybe creating a new ablp_loader utility file? Might help make our code a bit easier to read/follow-through s.t. if users are clicking through and trying to understand what/how to use ABLP loader at a high level, they don't have to be exposed to this internal label mapping logic as the first thing they see.

mkolodner-sc · 2026-07-27T10:50:52Z

+        Every anchor ``0..num_anchors-1`` receives a key; anchors with no labels
+        map to an empty ``long`` tensor on the same device as ``label_index``.


Can we add a note here that this we may eventually want to remove this path in the future?

mkolodner-sc · 2026-07-27T10:52:39Z

        # Set self._shutdowned right away, that way if we throw here, and __del__ is called,
        # then we can properly clean up and don't get extraneous error messages.
        self._shutdowned = True
+        self._use_list_output = use_list_output


Can we add a warning here that this will be deprecated in a future release? Unless you see some benefit to us keeping this around.

mkolodner-sc · 2026-07-27T11:04:47Z

+
+    num_labels = int(label_tensor.size(1))
+    flat = label_tensor.reshape(-1)  # [N_anchors * M] before the pad mask
+    # step 0 (see docstring example): tag each flattened entry with its anchor row.


Similarly, in-line comments are useful, but these feel a bit too long and hard to follow for me (we have step 0, step 0 continued, precondition for step 1, step 1 ...)

mkolodner-sc · 2026-07-27T11:15:02Z

+    """
+    sorted_cache: dict[NodeType, tuple[torch.Tensor, torch.Tensor]] = {}
+
+    def _sorted_for(node_type: NodeType) -> tuple[torch.Tensor, torch.Tensor]:


sorted_for is a bit ambiguous of a name, can we make this clearer? Ditto for the other functions here (i.e. "remap" is also vague.

mkolodner-sc · 2026-07-27T11:17:13Z

+    # searchsorted returns N_nodes for an id larger than every entry, which would
+    # gather out of bounds at step 2; clamp it back into range.
+    sorted_positions = sorted_positions.clamp_(max=num_nodes - 1)


In what scenarios do we expect this to happen?

mkolodner-sc · 2026-07-27T11:18:43Z

+        anchor_of_matched.to(to_device).to(torch.long),
+        local_index.to(to_device).to(torch.long),


Do we need .to(to_device).to(torch.long)? Asking since we also call to_device in the training example loops

mkolodner-sc · 2026-07-27T11:22:12Z

+    )
+
+
+def edge_list_set_labels(


nit: I think some of the new function and var names introduced in this PR are a bit non-intuitive for me. A few examples

edge_list_set_labels might be easier to make sense of if we put the action first i.e. set_edge_list_labels

node_local_to_global_by_type could be better and more aligned with other vars as local_id_to_global_id_by_node_type

sorted_node and sorted_perm could have easier names to follow as well

kmontemayor2-sc · 2026-07-28T16:48:21Z

/all_test

github-actions · 2026-07-28T16:48:34Z

GiGL Automation

@ 16:48:34UTC : 🔄 Integration Test started.

@ 18:05:25UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T16:48:35Z

GiGL Automation

@ 16:48:35UTC : 🔄 Lint Test started.

@ 16:56:39UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T16:48:38Z

GiGL Automation

@ 16:48:38UTC : 🔄 E2E Test started.

@ 18:17:36UTC : ❌ Workflow failed.
Please check the logs for more details.

github-actions · 2026-07-28T16:48:39Z

GiGL Automation

@ 16:48:39UTC : 🔄 Python Unit Test started.

@ 17:56:56UTC : ❌ Workflow failed.
Please check the logs for more details.

github-actions · 2026-07-28T16:48:41Z

GiGL Automation

@ 16:48:41UTC : 🔄 C++ Unit Test started.

@ 16:50:36UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T16:48:42Z

GiGL Automation

@ 16:48:42UTC : 🔄 Scala Unit Test started.

@ 16:57:26UTC : ✅ Workflow completed successfully.

…dge types _set_labels built the local->global node map from self._supervision_edge_types[2]. Those edge types are stored reversed when edge_dir="in" (dist_ablp_neighborloader.py:486-490), so that index is the anchor node type there, while the label edge types reaching collation stay outward-facing. Every heterogeneous edge_dir="in" batch therefore raised KeyError on its supervision node type. The pre-refactor code inserted both endpoints of each supervision edge type, which masked the direction ambiguity. Deriving the map from the label edge type keys that _remap_labels_by_edge_type actually looks up drops the edge_dir coupling entirely. Add heterogeneous coverage for both output formats across both edge dirs. The existing equivalence test is homogeneous and edge_dir="out", so it could not catch this. Anchors and supervision nodes use distinct node types, so a label remapped against the wrong node map yields a wrong global id instead of passing silently. Fixes the 5 unit errors and the het-dblp-sup{,-gs} e2e trainer failures in https://github.com/Snapchat/GiGL/actions/runs/30379971583 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

kmontemayor2-sc · 2026-07-28T23:42:06Z

/all_test

github-actions · 2026-07-28T23:42:17Z

GiGL Automation

@ 23:42:16UTC : 🔄 Python Unit Test started.

@ 01:01:08UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T23:42:19Z

GiGL Automation

@ 23:42:18UTC : 🔄 Scala Unit Test started.

@ 23:51:08UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T23:42:19Z

GiGL Automation

@ 23:42:19UTC : 🔄 Lint Test started.

@ 23:51:09UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T23:42:21Z

GiGL Automation

@ 23:42:21UTC : 🔄 C++ Unit Test started.

@ 23:44:57UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T23:42:23Z

GiGL Automation

@ 23:42:22UTC : 🔄 E2E Test started.

@ 01:07:18UTC : ✅ Workflow completed successfully.

github-actions · 2026-07-28T23:42:27Z

GiGL Automation

@ 23:42:26UTC : 🔄 Integration Test started.

@ 01:05:49UTC : ✅ Workflow completed successfully.

kmontemayor and others added 12 commits June 25, 2026 14:57

feat(ablp): add vectorized_set_labels kernel, equivalence-tested vs l…

00f4cdb

…oop oracle Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

test(ablp): add CUDA device-placement regression for label-remap kernels

3101506

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

feat(ablp): vectorized label remap always + use_list_output ctor flag

d51bd7a

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docs(ablp): document vectorized remap, use_list_output, and AnchorLabels

efa446d

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docs(examples): consume AnchorLabels edge-list in homogeneous training

8c91d66

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docs(examples): consume AnchorLabels edge-list in heterogeneous training

1f1c069

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docs(examples): consume AnchorLabels edge-list in graph-store homogen…

f60df78

…eous training Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docs(examples): consume AnchorLabels edge-list in graph-store heterog…

ee4adeb

…eneous training Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

feat(distributed): export AnchorLabels from gigl.distributed

dab9f8f

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

kmontemayor2-sc force-pushed the kmonte/ablp-vectorized-labels-and-list-output branch from 292f4d4 to bd97d03 Compare June 25, 2026 14:57

kmontemayor and others added 2 commits June 25, 2026 16:18

kmontemayor2-sc force-pushed the kmonte/ablp-vectorized-labels-and-list-output branch from 486e670 to d42d0df Compare June 26, 2026 19:24

kmontemayor2-sc force-pushed the kmonte/ablp-vectorized-labels-and-list-output branch from d42d0df to 641b24e Compare June 26, 2026 21:13

kmontemayor and others added 3 commits June 26, 2026 21:47

kmontemayor2-sc changed the title ~~Kmonte/ablp vectorized labels and list output~~ DistABLPLoader: vectorized label remap + optional AnchorLabels edge-list output Jun 29, 2026

mkolodner-sc reviewed Jul 27, 2026

View reviewed changes

refactor: represent ABLP labels as edge index tensors

b86604d

		num_anchors (int): Total number of anchors ``N_anchors`` (rows of the
		source padded label tensor), including anchors with no labels.

		anchor_index (torch.Tensor): ``[E]`` long tensor of local anchor rows.
		label_index (torch.Tensor): ``[E]`` long tensor of local label node ids.

		Every anchor ``0..num_anchors-1`` receives a key; anchors with no labels
		map to an empty ``long`` tensor on the same device as ``label_index``.

		anchor_of_matched.to(to_device).to(torch.long),
		local_index.to(to_device).to(torch.long),

Uh oh!

Conversation

kmontemayor2-sc commented Jun 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Tests

Notes

Uh oh!

mkolodner-sc left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kmontemayor2-sc commented Jul 28, 2026

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

kmontemayor2-sc commented Jul 28, 2026

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

GiGL Automation

Uh oh!

github-actions Bot commented Jul 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

kmontemayor2-sc commented Jun 25, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading

github-actions Bot commented Jul 28, 2026 •

edited

Loading