forked from onnx/optimizer
-
Notifications
You must be signed in to change notification settings - Fork 0
Pull requests: onnxsim/optimizer
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
extract_constant_to_initializer: batch-reserve names instead of one scan per match
#12
opened Aug 20, 2026 by
take-cheeze
Member
Loading…
3 tasks done
eliminate_deadend/eliminate_common_subexpression: avoid uses()'s O(N) capture scan
#11
opened Aug 20, 2026 by
take-cheeze
Member
Loading…
3 tasks done
Cache CSETensorHash's raw_data hash by tensor_id(), the actual bottleneck
#10
opened Aug 19, 2026 by
take-cheeze
Member
Loading…
Update Tensor* users for Graph::initializers_'s new unique_ptr storage
#9
opened Aug 19, 2026 by
take-cheeze
Member
Loading…
Avoid ParseTensorData's double copy in CSE tensor hash/equality for raw_data
#8
opened Aug 18, 2026 by
take-cheeze
Member
Loading…
Fusions for onnxsim node-count parity: ConvTranspose, no-op Dropout, Unsqueeze & batched Gemm
#7
opened Jul 31, 2026 by
take-cheeze
Member
Loading…
Add fuse_matmul_add_bias_into_gemm_batched pass
#6
opened Jul 29, 2026 by
take-cheeze
Member
Loading…
Add option to treat initializers as non-constants
#2
opened Jul 29, 2026 by
take-cheeze
Member
Loading…
ProTip!
Follow long discussions with comments:>50.