Python torch.nn.parallel.parallel_apply.parallel_apply() Examples

The following are 30 code examples of torch.nn.parallel.parallel_apply.parallel_apply(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module torch.nn.parallel.parallel_apply , or try the search function .
Example #1
Source File: EncodingDataParallel.py    From torch-toolbox with BSD 3-Clause "New" or "Revised" License 6 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)

        for t in chain(self.module.parameters(), self.module.buffers()):
            if t.device != self.src_device_obj:
                raise RuntimeError(
                    "module must have its parameters and buffers "
                    "on device {} (device_ids[0]) but found one of "
                    "them on device: {}".format(
                        self.src_device_obj, t.device))
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids)
        if len(self.device_ids) == 1:
            return self.module(*inputs, **kwargs)
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return outputs
Example #2
Source File: data_parallel.py    From centerNet-deep-sort with GNU General Public License v3.0 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #3
Source File: data_parallel.py    From CenterNet-CondInst with MIT License 5 votes vote down vote up 
def data_parallel(module, inputs, device_ids=None, output_device=None, dim=0, module_kwargs=None):
    r"""Evaluates module(input) in parallel across the GPUs given in device_ids.

    This is the functional version of the DataParallel module.

    Args:
        module: the module to evaluate in parallel
        inputs: inputs to the module
        device_ids: GPU ids on which to replicate module
        output_device: GPU location of the output  Use -1 to indicate the CPU.
            (default: device_ids[0])
    Returns:
        a Variable containing the result of module(input) located on
        output_device
    """
    if not isinstance(inputs, tuple):
        inputs = (inputs,)

    if device_ids is None:
        device_ids = list(range(torch.cuda.device_count()))

    if output_device is None:
        output_device = device_ids[0]

    inputs, module_kwargs = scatter_kwargs(inputs, module_kwargs, device_ids, dim)
    if len(device_ids) == 1:
        return module(*inputs[0], **module_kwargs[0])
    used_device_ids = device_ids[:len(inputs)]
    replicas = replicate(module, used_device_ids)
    outputs = parallel_apply(replicas, inputs, module_kwargs, used_device_ids)
    return gather(outputs, output_device, dim)
Example #4
Source File: data_parallel.py    From BCAI_kaggle_CHAMPS with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        if self.gpu0_bsz == 0:
            device_ids = self.device_ids[1:]
        else:
            device_ids = self.device_ids
        inputs, kwargs = self.scatter(inputs, kwargs, device_ids)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids)
        if self.gpu0_bsz == 0:
            replicas = replicas[1:]
        outputs = self.parallel_apply(replicas, device_ids, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #5
Source File: data_parallel.py    From BCAI_kaggle_CHAMPS with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, device_ids, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, device_ids)
Example #6
Source File: data_parallel.py    From deq with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        if self.gpu0_bsz == 0:
            device_ids = self.device_ids[1:]
        else:
            device_ids = self.device_ids
        inputs, kwargs = self.scatter(inputs, kwargs, device_ids)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids)
        if self.gpu0_bsz == 0:
            replicas = replicas[1:]
        outputs = self.parallel_apply(replicas, device_ids, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #7
Source File: data_parallel.py    From deq with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, device_ids, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, device_ids)
Example #8
Source File: data_parallel.py    From CornerNet with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #9
Source File: data_parallel.py    From CornerNet with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #10
Source File: data_parallel.py    From CornerNet with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def data_parallel(module, inputs, device_ids=None, output_device=None, dim=0, module_kwargs=None):
    r"""Evaluates module(input) in parallel across the GPUs given in device_ids.

    This is the functional version of the DataParallel module.

    Args:
        module: the module to evaluate in parallel
        inputs: inputs to the module
        device_ids: GPU ids on which to replicate module
        output_device: GPU location of the output  Use -1 to indicate the CPU.
            (default: device_ids[0])
    Returns:
        a Variable containing the result of module(input) located on
        output_device
    """
    if not isinstance(inputs, tuple):
        inputs = (inputs,)

    if device_ids is None:
        device_ids = list(range(torch.cuda.device_count()))

    if output_device is None:
        output_device = device_ids[0]

    inputs, module_kwargs = scatter_kwargs(inputs, module_kwargs, device_ids, dim)
    if len(device_ids) == 1:
        return module(*inputs[0], **module_kwargs[0])
    used_device_ids = device_ids[:len(inputs)]
    replicas = replicate(module, used_device_ids)
    outputs = parallel_apply(replicas, inputs, module_kwargs, used_device_ids)
    return gather(outputs, output_device, dim)
Example #11
Source File: data_parallel.py    From centerNet-deep-sort with GNU General Public License v3.0 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #12
Source File: data_parallel.py    From CenterNet-CondInst with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #13
Source File: data_parallel.py    From centerNet-deep-sort with GNU General Public License v3.0 5 votes vote down vote up 
def data_parallel(module, inputs, device_ids=None, output_device=None, dim=0, module_kwargs=None):
    r"""Evaluates module(input) in parallel across the GPUs given in device_ids.

    This is the functional version of the DataParallel module.

    Args:
        module: the module to evaluate in parallel
        inputs: inputs to the module
        device_ids: GPU ids on which to replicate module
        output_device: GPU location of the output  Use -1 to indicate the CPU.
            (default: device_ids[0])
    Returns:
        a Variable containing the result of module(input) located on
        output_device
    """
    if not isinstance(inputs, tuple):
        inputs = (inputs,)

    if device_ids is None:
        device_ids = list(range(torch.cuda.device_count()))

    if output_device is None:
        output_device = device_ids[0]

    inputs, module_kwargs = scatter_kwargs(inputs, module_kwargs, device_ids, dim)
    if len(device_ids) == 1:
        return module(*inputs[0], **module_kwargs[0])
    used_device_ids = device_ids[:len(inputs)]
    replicas = replicate(module, used_device_ids)
    outputs = parallel_apply(replicas, inputs, module_kwargs, used_device_ids)
    return gather(outputs, output_device, dim)
Example #14
Source File: data_parallel_dist.py    From ps_pytorch with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids)
        self._sync_params()
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        outputs = self.parallel_apply(self._module_copies, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #15
Source File: data_parallel_dist.py    From ps_pytorch with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #16
Source File: parallel.py    From kekas with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[: len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return outputs
Example #17
Source File: parallel.py    From kekas with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[: len(replicas)])
Example #18
Source File: parallel.py    From kekas with MIT License 5 votes vote down vote up 
def forward(self, inputs, *targets, **kwargs):
        # input should be already scattered
        # scattering the targets instead
        targets, kwargs = self.scatter(targets, kwargs, self.device_ids)
        if len(self.device_ids) == 1:
            return self.module(inputs, *targets[0], **kwargs[0])
        inputs = [(input,) for input in inputs]
        replicas = self.replicate(self.module, self.device_ids[: len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, targets, kwargs)
        return self.gather(outputs, self.output_device)
Example #19
Source File: parallel.py    From kekas with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, targets, kwargs):
        return criterion_parallel_apply(
            replicas, inputs, targets, kwargs, self.device_ids[: len(replicas)]
        )
Example #20
Source File: data_parallel.py    From CenterNet with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #21
Source File: data_parallel.py    From CenterNet with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #22
Source File: data_parallel.py    From CenterNet with MIT License 5 votes vote down vote up 
def data_parallel(module, inputs, device_ids=None, output_device=None, dim=0, module_kwargs=None):
    r"""Evaluates module(input) in parallel across the GPUs given in device_ids.

    This is the functional version of the DataParallel module.

    Args:
        module: the module to evaluate in parallel
        inputs: inputs to the module
        device_ids: GPU ids on which to replicate module
        output_device: GPU location of the output  Use -1 to indicate the CPU.
            (default: device_ids[0])
    Returns:
        a Variable containing the result of module(input) located on
        output_device
    """
    if not isinstance(inputs, tuple):
        inputs = (inputs,)

    if device_ids is None:
        device_ids = list(range(torch.cuda.device_count()))

    if output_device is None:
        output_device = device_ids[0]

    inputs, module_kwargs = scatter_kwargs(inputs, module_kwargs, device_ids, dim)
    if len(device_ids) == 1:
        return module(*inputs[0], **module_kwargs[0])
    used_device_ids = device_ids[:len(inputs)]
    replicas = replicate(module, used_device_ids)
    outputs = parallel_apply(replicas, inputs, module_kwargs, used_device_ids)
    return gather(outputs, output_device, dim)
Example #23
Source File: data_parallel.py    From CenterNet with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #24
Source File: EncodingDataParallel.py    From torch-toolbox with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs,
                              self.device_ids[:len(replicas)])
Example #25
Source File: data_parallel.py    From CornerNet-Lite-Pytorch with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        # 分散输入到各个设备里
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        # 复制啥啊
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        # 使用并行函数获得输出
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)

    # 对torch里的函数进行了再包装
Example #26
Source File: data_parallel.py    From CornerNet-Lite-Pytorch with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def parallel_apply(self, replicas, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
Example #27
Source File: data_parallel.py    From CornerNet-Lite-Pytorch with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def data_parallel(module, inputs, device_ids=None, output_device=None, dim=0, module_kwargs=None):
    r"""Evaluates module(input) in parallel across the GPUs given in device_ids.

    This is the functional version of the DataParallel module.

    Args:
        module: the module to evaluate in parallel
        inputs: inputs to the module
        device_ids: GPU ids on which to replicate module
        output_device: GPU location of the output  Use -1 to indicate the CPU.
            (default: device_ids[0])
    Returns:
        a Variable containing the result of module(input) located on
        output_device
    """
    if not isinstance(inputs, tuple):
        inputs = (inputs,)

    if device_ids is None:
        device_ids = list(range(torch.cuda.device_count()))

    if output_device is None:
        output_device = device_ids[0]

    inputs, module_kwargs = scatter_kwargs(inputs, module_kwargs, device_ids, dim)
    if len(device_ids) == 1:
        return module(*inputs[0], **module_kwargs[0])
    used_device_ids = device_ids[:len(inputs)]
    replicas = replicate(module, used_device_ids)
    outputs = parallel_apply(replicas, inputs, module_kwargs, used_device_ids)
    return gather(outputs, output_device, dim)
Example #28
Source File: data_parallel.py    From FEAT with MIT License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        if self.gpu0_bsz == 0:
            device_ids = self.device_ids[1:]
        else:
            device_ids = self.device_ids
        inputs, kwargs = self.scatter(inputs, kwargs, device_ids)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids)
        if self.gpu0_bsz == 0:
            replicas = replicas[1:]
        outputs = self.parallel_apply(replicas, device_ids, inputs, kwargs)
        return self.gather(outputs, self.output_device)
Example #29
Source File: data_parallel.py    From FEAT with MIT License 5 votes vote down vote up 
def parallel_apply(self, replicas, device_ids, inputs, kwargs):
        return parallel_apply(replicas, inputs, kwargs, device_ids)
Example #30
Source File: data_parallel.py    From ExtremeNet with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up 
def forward(self, *inputs, **kwargs):
        if not self.device_ids:
            return self.module(*inputs, **kwargs)
        inputs, kwargs = self.scatter(inputs, kwargs, self.device_ids, self.chunk_sizes)
        if len(self.device_ids) == 1:
            return self.module(*inputs[0], **kwargs[0])
        replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
        outputs = self.parallel_apply(replicas, inputs, kwargs)
        return self.gather(outputs, self.output_device)