Refactor tensor initialization check for external weights and fixed lint issues#436
Refactor tensor initialization check for external weights and fixed lint issues#436jatinwadhwa921 wants to merge 22 commits intoovep-develop-lnl-1.2from
Conversation
fix handling of multiple QuantizeLinear nodes
fix debug build
* Update onnx_ctx_model_helper.cc Keep "attr_0" "attr_1" etc inside a scope so that these unique pointers get deleted after their use is over Also declare "node_attributes" global to the new scope (within the function) * Optimizations to reduce memory used during compilation * Fix for reusing the serialized compiled blob to proto model * EpCtx export code refactoring * Fix from Ankit to move memory during ONNX Graph creation * Localise the OV IR model in memory * Apply lint changes --------- Co-authored-by: Vishnudas Thaniel S <vishnudas.thaniel.s@intel.com> Co-authored-by: Javier E. Martinez <javier.e.martinez@intel.com>
Updated the Version of OV to 2024.3
update: Enable C++ 20 Standard for the build
Reduced model proto scope
Upgrade Openvino version to 2024.3.0
Fix OVEP execution failure with dynamic shaped models
jatinwadhwa921
requested review from
ankitm3k,
preetha-intel,
saurabhkale17,
sfatimar and
vthaniel
|
LGTM |
preetha-intel
left a comment
preetha-intel
left a comment
There was a problem hiding this comment.
Passing this variable across call stack might be difficult to trace in long term. Can you explore the ng_required_initialisers in capability.cc or data_ops.cc
saurabhkale17
force-pushed
the
jatin/external_weights
branch
from
67be471 to
89d0b7c
Compare
External weights can be associated with any initializer in the ONNX model, so we can miss the initializer if we are just traversing the nodes which are not supported as ng_required_initilisers points to the nodes which are not supported, so i believe we should go ahead with my method for now |
|
LGTM |
sfatimar
force-pushed
the
ovep-develop-lnl-1.2
branch
from
236df14 to
b0a8bee
Compare
saurabhkale17
force-pushed
the
ovep-develop-lnl-1.2
branch
from
4d64dc0 to
f9b995c
Compare
|
@jatinwadhwa921 Rebase with latest changes. |
|
PR #464 is merged . Closing this as its no longer relevant. |
7 participants
Refactor tensor initialization check for external weights and fixed lint issues