Open
Conversation
- Introduced support for multiple Runtime instances - Added utility functions for multi-runtime management - Ensured backward compatibility with existing usage patterns of Runtime - Added unit tests for multi-runtime functionality --------- Co-authored-by: noli <t-ngerawork@microsoft.com>
- Adds Torch to ARK tensor conversion support - New ModelBufferManager class handles external buffer registration and simplifies buffer access during kernel initialization - Adds test cases for ARK to Torch conversion support --------- Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
…o chhwang/python
Merge main (including PRs #222, #235, #245) into the PR #215 branch. Resolution strategy: - Take main's version for all C++ core, Python API, and C++ binding files (features from PR #215 were reworked and landed via PRs #222 and #235) - Remove superseded PR #215 files: - torch_mock.py (replaced by python/ark/torch/mock.py) - unittest_common.py (replaced by python/unittest/common.py) - model_buffer_manager.hpp (replaced by buffer_registry) - arkprof.py (Profiler class in main suffices) - model_7b_b1_s2048.py / plan_llama2_7b_b1_s2048.json (superseded by current llama examples) - Rewrite test_conversion.py for current API: - get_torch_view() -> to_torch() - Remove delete_all_runtimes() / reset() calls - Use pytest_ark decorator and with-block pattern - Rewrite torch_tutorial.py to use placeholder API instead of RuntimeModule - Add test_conversion to test runner
chhwang
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Python API
Tensor.eval(stream=None)— evaluate graph and return torch tensortorch.Tensordirectly (auto-conversion)set_model(),current_model(),use_model()Planner(model=...)parameterC++ fixes
Tileconfig for correct grid alignment when fusedWwiseReduce: per-row reduction fix for multi-row tilesTests
test_eval.py: caching, recompilation, stream interleaving, chained opsExamples
eval()+ torch tensors