Skip to content

[hotfix]: remove unused code and deprecate colo_init_context #5117

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cwher merged 2 commits into from
Nov 28, 2023
Merged

[hotfix]: remove unused code and deprecate colo_init_context #5117

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e429d5e
fix: remove unused code
cwher Nov 28, 2023
81fbec2
feat: move colo_init_context to legacy
cwher Nov 28, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions colossalai/legacy/zero/gemini/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@
from .colo_init_context import ColoInitContext, post_process_colo_init_ctx
from .ophooks import BaseOpHook, register_ophooks_recursively
from .stateful_tensor import StatefulTensor
from .stateful_tensor_mgr import StatefulTensorMgr
Expand All @@ -11,4 +12,6 @@
"AutoTensorPlacementPolicy",
"register_ophooks_recursively",
"BaseOpHook",
"ColoInitContext",
"post_process_colo_init_ctx",
]
0 colossalai/zero/gemini/colo_init_context.py → ...i/legacy/zero/gemini/colo_init_context.py
View file
Open in desktop
File renamed without changes.
1 change: 0 additions & 1 deletion colossalai/shardformer/policies/gptj.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -163,7 +163,6 @@ def get_held_layers(self) -> List[nn.Module]:
layers_per_stage = self.distribute_layers(len(module.h), stage_manager.num_stages)
if stage_manager.is_first_stage():
held_layers.append(module.wte)
# held_layers.append(module.wpe)
held_layers.append(module.drop)
start_idx, end_idx = self.get_stage_index(layers_per_stage, stage_manager.stage)
held_layers.extend(module.h[start_idx:end_idx])
Expand Down
11 changes: 1 addition & 10 deletions colossalai/zero/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,4 @@
from .gemini import (
ColoInitContext,
GeminiAdamOptimizer,
GeminiDDP,
GeminiOptimizer,
get_static_torch_model,
post_process_colo_init_ctx,
)
from .gemini import GeminiAdamOptimizer, GeminiDDP, GeminiOptimizer, get_static_torch_model
from .low_level import LowLevelZeroOptimizer
from .wrapper import zero_model_wrapper, zero_optim_wrapper

Expand All @@ -16,7 +9,5 @@
"zero_model_wrapper",
"zero_optim_wrapper",
"LowLevelZeroOptimizer",
"ColoInitContext",
"post_process_colo_init_ctx",
"get_static_torch_model",
]
3 changes: 0 additions & 3 deletions colossalai/zero/gemini/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
from .chunk import ChunkManager, TensorInfo, TensorState, search_chunk_configuration
from .colo_init_context import ColoInitContext, post_process_colo_init_ctx
from .gemini_ddp import GeminiDDP
from .gemini_mgr import GeminiManager
from .gemini_optimizer import GeminiAdamOptimizer, GeminiOptimizer
Expand All @@ -15,6 +14,4 @@
"get_static_torch_model",
"GeminiAdamOptimizer",
"GeminiOptimizer",
"ColoInitContext",
"post_process_colo_init_ctx",
]