tensordict.nn.distributions.NormalParamExtractor¶
- class tensordict.nn.distributions.NormalParamExtractor(scale_mapping: str = 'biased_softplus_1.0', scale_lb: Number = 0.0001)¶
一个非参数化的 nn.Module,它将其输入拆分为 loc 和 scale 参数。
scale 参数使用指定的
scale_mapping
映射到正值。- 参数:
scale_mapping (str, 可选) – 与 std 一起使用的正映射函数。 默认值 =
"biased_softplus_1.0"
(即 softplus 映射,带有偏差,使得 fn(0.0) = 1.0) 选项:"softplus"
,"exp"
,"relu"
,"biased_softplus_1"
或"none"
(无映射)。有关更多详细信息,请参阅mappings()
。scale_lb (Number, 可选) – 方差可以取的最小值。默认值为 1e-4。
示例
>>> import torch >>> from tensordict.nn.distributions import NormalParamExtractor >>> from torch import nn >>> module = nn.Linear(3, 4) >>> normal_params = NormalParamExtractor() >>> tensor = torch.randn(3) >>> loc, scale = normal_params(module(tensor)) >>> print(loc.shape, scale.shape) torch.Size([2]) torch.Size([2]) >>> assert (scale > 0).all() >>> # with modules that return more than one tensor >>> module = nn.LSTM(3, 4) >>> tensor = torch.randn(4, 2, 3) >>> loc, scale, others = normal_params(*module(tensor)) >>> print(loc.shape, scale.shape) torch.Size([4, 2, 2]) torch.Size([4, 2, 2]) >>> assert (scale > 0).all()