mirror of https://github.com/hpcaitech/ColossalAI
30 lines
962 B
Python
30 lines
962 B
Python
import torch
|
|
from colossalai.tensor.op_wrapper import colo_op_impl
|
|
from colossalai.tensor import ColoTensor
|
|
|
|
|
|
@colo_op_impl(torch.mean)
|
|
def colo_mean(types, args=(), kwargs=None, pg=None):
|
|
stateful_tensor = args[0]
|
|
return torch.mean(stateful_tensor.torch_tensor())
|
|
|
|
|
|
def register_elementwise_op(op):
|
|
|
|
@colo_op_impl(op)
|
|
def elementwise_op(types, args=(), kwargs=None, pg=None):
|
|
"""
|
|
Handles ``__torch_function__`` dispatch for the elementwise op such
|
|
as ``torch.nn.functional.gelu`` or ``torch.nn.functional.relu``.
|
|
This method computes on either a normal tensor or a sharded tensor.
|
|
"""
|
|
input_tensor = args[0]
|
|
# Validate types
|
|
if not isinstance(input_tensor, ColoTensor):
|
|
raise TypeError("input needs to be a ColoTensor")
|
|
return op(input_tensor.torch_tensor())
|
|
|
|
|
|
register_elementwise_op(torch.nn.functional.gelu)
|
|
register_elementwise_op(torch.nn.functional.relu)
|