本文整理汇总了Python中torch.cuda.comm.reduce_add_coalesced方法的典型用法代码示例。如果您正苦于以下问题：Python comm.reduce_add_coalesced方法的具体用法？Python comm.reduce_add_coalesced怎么用？Python comm.reduce_add_coalesced使用的例子？那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类torch.cuda.comm的用法示例。
在下文中一共展示了comm.reduce_add_coalesced方法的5个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: forward
# 需要导入模块: from torch.cuda import comm [as 别名]
# 或者: from torch.cuda.comm import reduce_add_coalesced [as 别名]
def forward(ctx, num_inputs, *inputs):
ctx.num_inputs = num_inputs
ctx.target_gpus = [inputs[i].get_device() for i in range(0, len(inputs), num_inputs)]
inputs = [inputs[i:i + num_inputs]
for i in range(0, len(inputs), num_inputs)]
# sort before reduce sum
inputs = sorted(inputs, key=lambda i: i[0].get_device())
results = comm.reduce_add_coalesced(inputs, ctx.target_gpus[0])
outputs = comm.broadcast_coalesced(results, ctx.target_gpus)
return tuple([t for tensors in outputs for t in tensors]) 示例2: backward
# 需要导入模块: from torch.cuda import comm [as 别名]
# 或者: from torch.cuda.comm import reduce_add_coalesced [as 别名]
def backward(ctx, *inputs):
inputs = [i.data for i in inputs]
inputs = [inputs[i:i + ctx.num_inputs]
for i in range(0, len(inputs), ctx.num_inputs)]
results = comm.reduce_add_coalesced(inputs, ctx.target_gpus[0])
outputs = comm.broadcast_coalesced(results, ctx.target_gpus)
return (None,) + tuple([Variable(t) for tensors in outputs for t in tensors]) 示例3: forward
# 需要导入模块: from torch.cuda import comm [as 别名]
# 或者: from torch.cuda.comm import reduce_add_coalesced [as 别名]
def forward(ctx, num_inputs, *inputs):
ctx.num_inputs = num_inputs
ctx.target_gpus = [inputs[i].get_device() for i in range(0, len(inputs), num_inputs)]
inputs = [inputs[i:i + num_inputs]
for i in range(0, len(inputs), num_inputs)]
# sort before reduce sum
inputs = sorted(inputs, key=lambda i: i[0].get_device())
results = comm.reduce_add_coalesced(inputs, ctx.target_gpus[0])
outputs = comm.broadcast_coalesced(results, ctx.target_gpus)
return tuple([t for tensors in outputs for t in tensors]) 示例4: backward
# 需要导入模块: from torch.cuda import comm [as 别名]
# 或者: from torch.cuda.comm import reduce_add_coalesced [as 别名]
def backward(ctx, *inputs):
inputs = [i.data for i in inputs]
inputs = [inputs[i:i + ctx.num_inputs]
for i in range(0, len(inputs), ctx.num_inputs)]
results = comm.reduce_add_coalesced(inputs, ctx.target_gpus[0])
outputs = comm.broadcast_coalesced(results, ctx.target_gpus)
return (None,) + tuple([Variable(t) for tensors in outputs for t in tensors]) 示例5: forward
# 需要导入模块: from torch.cuda import comm [as 别名]
# 或者: from torch.cuda.comm import reduce_add_coalesced [as 别名]
def forward(ctx, destination, num_inputs, *grads):
ctx.target_gpus = [grads[i].get_device() for i in range(0, len(grads), num_inputs)]
grads = [grads[i:i + num_inputs]
for i in range(0, len(grads), num_inputs)]
return comm.reduce_add_coalesced(grads, destination)