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.
ColossalAI/colossalai/utils/commons/bucket_tensor_copy.py

62 lines
1.9 KiB

import torch
from colossalai.zero.sharded_param import ShardedParamV2
from colossalai.utils import get_current_device
from typing import List
class BucketizedTensorCopy(object):
def __init__(
self,
chunk_size: int,
):
r"""
torch.nn.Parameter CPU (fp32) -> ShardedParam GPU (fp16)
TODO(jiaruifang) The class is a little bit hardcoded
I will make it more general later.
"""
self.chunk_size = chunk_size
self._offset = 0
self._cpu_buffer = torch.empty(chunk_size, dtype=torch.float, device=torch.device("cpu:0"), pin_memory=True)
self._cuda_buffer = torch.empty(chunk_size,
dtype=torch.half,
device=torch.device(f"cuda:{get_current_device()}"))
self._buffered_param_list: List[ShardedParamV2] = []
self._numel_list = []
def copy(self, src_param: torch.nn.Parameter, target_param: ShardedParamV2):
assert isinstance(target_param, ShardedParamV2)
assert isinstance(src_param, torch.nn.Parameter)
numel = src_param.numel()
if self._offset + numel > self.chunk_size:
self.flush()
assert src_param.data.device.type == 'cpu'
self._cpu_buffer.narrow(0, self._offset, numel).copy_(src_param.data.view(-1))
self._buffered_param_list.append(target_param)
self._numel_list.append(numel)
self._offset += numel
def flush(self):
"""
flush to cuda memory
"""
self._cuda_buffer.copy_(self._cpu_buffer)
flush_offset = 0
for sparam, numel in zip(self._buffered_param_list, self._numel_list):
sparam.sharded_data_tensor.copy_payload(self._cpu_buffer.narrow(0, flush_offset, numel))
flush_offset += numel
self.reset()
def reset(self):
self._buffered_param_list = []
self._numel_list = []
self._offset = 0