mirror of https://github.com/hpcaitech/ColossalAI
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
98 lines
3.4 KiB
98 lines
3.4 KiB
from contextlib import contextmanager
|
|
from enum import Enum
|
|
from functools import partial
|
|
from typing import List
|
|
|
|
import torch
|
|
|
|
from colossalai.tensor.param_op_hook import ColoParamOpHook
|
|
from colossalai.utils import is_ddp_ignored
|
|
from colossalai.zero.gemini import TensorState
|
|
from colossalai.zero.gemini.gemini_mgr import GeminiManager
|
|
|
|
|
|
class TrainingPhase(Enum):
|
|
FORWARD = 0
|
|
BACKWARD = 1
|
|
|
|
|
|
class GeminiZeROHook(ColoParamOpHook):
|
|
def __init__(self, gemini_manager: GeminiManager) -> None:
|
|
super().__init__()
|
|
self._gemini_manager = gemini_manager
|
|
self._chunk_manager = gemini_manager.chunk_manager
|
|
self._training_phase = TrainingPhase.FORWARD
|
|
|
|
def pre_op(self, params):
|
|
# map params to chunks
|
|
params = [p for p in params if not is_ddp_ignored(p)]
|
|
all_chunks = self._chunk_manager.get_chunks(params)
|
|
|
|
# wait for prefetched chunks, filter those are not prefetched
|
|
chunks_fetch_sync = self._gemini_manager.wait_chunks(all_chunks)
|
|
|
|
# transfer state
|
|
for p in params:
|
|
self._chunk_manager.trans_tensor_state(p, TensorState.COMPUTE)
|
|
self._gemini_manager.sample_overall_data()
|
|
|
|
# evit chunks, aware of async fetched
|
|
self._gemini_manager.adjust_layout(
|
|
all_chunks, record_anyway=self._gemini_manager.placement_policy.max_prefetch > 0
|
|
)
|
|
|
|
# fetch the rest synchronously
|
|
for chunk in chunks_fetch_sync:
|
|
self._chunk_manager.access_chunk(chunk)
|
|
|
|
# get possible chunks to prefetch
|
|
chunks_fetch_async = self._gemini_manager.placement_policy.get_prefetch_chunks(
|
|
is_warmup=self._gemini_manager.is_warmup(),
|
|
compute_list=self._gemini_manager.compute_list,
|
|
compute_idx=self._gemini_manager.compute_idx,
|
|
async_works=self._gemini_manager.async_works,
|
|
)
|
|
|
|
# prefetch
|
|
for chunk in chunks_fetch_async:
|
|
maybe_work = self._chunk_manager.access_chunk(chunk, async_access=True)
|
|
if maybe_work is not None:
|
|
self._gemini_manager.add_work(chunk, maybe_work)
|
|
|
|
# record cuda model data of the current OP, including memory for prefetched chunks
|
|
self._gemini_manager.record_model_data_volume()
|
|
|
|
def post_op(self, params):
|
|
params = [p for p in params if not is_ddp_ignored(p)]
|
|
for p in params:
|
|
tensor_state = (
|
|
TensorState.HOLD
|
|
if self._training_phase == TrainingPhase.FORWARD or not p.requires_grad
|
|
else TensorState.HOLD_AFTER_BWD
|
|
)
|
|
self._chunk_manager.trans_tensor_state(p, tensor_state)
|
|
|
|
def pre_forward(self, params: List[torch.Tensor]) -> None:
|
|
self.pre_op(params)
|
|
|
|
def post_forward(self, params: List[torch.Tensor]) -> None:
|
|
self.post_op(params)
|
|
|
|
def pre_backward(self, params: List[torch.Tensor]) -> None:
|
|
self.pre_op(params)
|
|
|
|
def post_backward(self, params: List[torch.Tensor]) -> None:
|
|
self.post_op(params)
|
|
|
|
@contextmanager
|
|
def switch_training_phase(self, training_phase: TrainingPhase = TrainingPhase.BACKWARD):
|
|
old_training_phase = self._training_phase
|
|
try:
|
|
self._training_phase = training_phase
|
|
yield
|
|
finally:
|
|
self._training_phase = old_training_phase
|
|
|
|
switch_to_backward = switch_training_phase
|
|
switch_to_forward = partial(switch_to_backward, training_phase=TrainingPhase.FORWARD)
|