Python torch.backends.cudnn 模块,FilterDescriptor() 实例源码

我们从Python开源项目中,提取了以下5个代码示例,用于说明如何使用torch.backends.cudnn.FilterDescriptor()

项目:pytorch-dist    作者:apaszke    | 项目源码 | 文件源码
def init_weight_descriptor(fn, weight):
    w_desc = cudnn.FilterDescriptor()
    w_view = weight.view(-1, 1, 1)  # seems that filters require >=3 dimensions
    w_desc.set(w_view)
    return w_desc
项目:pytorch    作者:tylergenter    | 项目源码 | 文件源码
def init_weight_descriptor(fn, weight):
    w_desc = cudnn.FilterDescriptor()
    w_view = weight.view(-1, 1, 1)  # seems that filters require >=3 dimensions
    w_desc.set(w_view)
    return w_desc
项目:pytorch-coriander    作者:hughperkins    | 项目源码 | 文件源码
def init_weight_descriptor(fn, weight):
    w_desc = cudnn.FilterDescriptor()
    w_view = weight.view(-1, 1, 1)  # seems that filters require >=3 dimensions
    w_desc.set(w_view)
    return w_desc
项目:pytorch    作者:ezyang    | 项目源码 | 文件源码
def init_weight_descriptor(fn, weight):
    w_desc = cudnn.FilterDescriptor()
    w_view = weight.view(-1, 1, 1)  # seems that filters require >=3 dimensions
    w_desc.set(w_view)
    return w_desc
项目:pytorch    作者:pytorch    | 项目源码 | 文件源码
def init_weight_descriptor(fn, weight):
    w_desc = cudnn.FilterDescriptor()
    w_view = weight.view(-1, 1, 1)  # seems that filters require >=3 dimensions
    w_desc.set(w_view)
    return w_desc