Enable passing initial optimizer state while creating training session #5869
Conversation
This file contains 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
ashbhandare
requested review from
BowenBao,
liqunfu,
spandantiwari,
thiagocrepaldi and
a team
as code owners
ashbhandare
force-pushed
the
aibhanda/load_optim_state
branch
from
3b17d48 to
0212536
Compare
orttraining/orttraining/test/graph/training_session_test_utils.h
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Nov 21, 2020
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/lamb_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/lamb_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
orttraining/orttraining/core/graph/optimizer/sgd_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
ashbhandare
force-pushed
the
aibhanda/load_optim_state
branch
4 times, most recently
from
d649a2a to
82376dc
Compare
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/core/graph/optimizer/adam_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/test/graph/optimizer_graph_builder_test.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/test/optimizer/horizontal_parallel_test_utils.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/test/session/training_session_test_utils.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
orttraining/orttraining/test/session/training_session_test_utils.h
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 2, 2020
edgchen1
reviewed
Dec 2, 2020
baijumeswani
reviewed
Dec 2, 2020
orttraining/orttraining/core/graph/optimizer/sgd_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 3, 2020
orttraining/orttraining/core/graph/optimizer/lamb_optimizer_builder.cc
Outdated
Show resolved
Hide resolved
edgchen1
reviewed
Dec 3, 2020
edgchen1
reviewed
Dec 3, 2020
edgchen1
reviewed
Dec 3, 2020
edgchen1
reviewed
Dec 3, 2020
baijumeswani
reviewed
Dec 4, 2020
orttraining/orttraining/test/graph/gradient_graph_builder_test.cc
Outdated
Show resolved
Hide resolved
baijumeswani
previously approved these changes
Dec 4, 2020
thiagocrepaldi
suggested changes
Dec 4, 2020
orttraining/orttraining/test/python/orttraining_test_orttrainer_bert_toy_onnx.py
Outdated
Show resolved
Hide resolved
orttraining/orttraining/test/python/orttraining_test_orttrainer_bert_toy_onnx.py
Outdated
Show resolved
Hide resolved
ashbhandare
force-pushed
the
aibhanda/load_optim_state
branch
from
8a8146b to
231a955
Compare
thiagocrepaldi
previously approved these changes
Dec 7, 2020
edgchen1
previously approved these changes
Dec 8, 2020
ashbhandare
dismissed stale reviews from edgchen1 and thiagocrepaldi
via
479a62b
ashbhandare
force-pushed
the
aibhanda/load_optim_state
branch
from
b8bed92 to
479a62b
Compare
thiagocrepaldi
approved these changes
Dec 8, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This change enables passing initial states for optimizers Adam and Lamb while creating training session. This is the first part of the changes required to enable loading optimizer state from a checkpoint into a model-parallel run(Zero/Megatron).
Also includes moving some common test functions to a separate training_session_test_utils.h.