Regression Metrics

Regression Metrics

参考地址：Module metrics — PyTorch-Metrics 0.4.0 documentation (torchmetrics.readthedocs.io)

CosineSimilarity

计算target和preds之间的CosineSimilarity公式：

where is a tensor of target values, and is a tensor of predictions.

Forward accepts

• preds (float tensor): (N,d)
• target (float tensor): (N,d)

>>> from torchmetrics import CosineSimilarity
>>> target = torch.tensor([[0, 1], [1, 1]])
>>> preds = torch.tensor([[0, 1], [0, 1]])
>>> cosine_similarity = CosineSimilarity(reduction = 'mean')
>>> cosine_similarity(preds, target)
tensor(0.8536)

ExplainedVariance

ExplainedVariance计算公式：

Where is a tensor of target values, and is a tensor of predictions.

Forward accepts

• preds (float tensor): (N,) or (N, ...) (multioutput)
• target (long tensor): (N,) or (N, ...) (multioutput)

在多输出的情况下，默认情况下，方差将在附加维度上均匀平均。 请参阅参数 multioutput 以更改此行为。

Example

>>>

>>> from torchmetrics import ExplainedVariance
>>> target = torch.tensor([3, -0.5, 2, 7])
>>> preds = torch.tensor([2.5, 0.0, 2, 8])
>>> explained_variance = ExplainedVariance()
>>> explained_variance(preds, target)
tensor(0.9572)

>>> target = torch.tensor([[0.5, 1], [-1, 1], [7, -6]])
>>> preds = torch.tensor([[0, 2], [-1, 2], [8, -5]])
>>> explained_variance = ExplainedVariance(multioutput='raw_values')
>>> explained_variance(preds, target)
tensor([0.9677, 1.0000])

MeanAbsoluteError

MeanAbsoluteError（MAE）计算公式：

Where is a tensor of target values, and is a tensor of predictions.

>>> from torchmetrics import MeanAbsoluteError
>>> target = torch.tensor([3.0, -0.5, 2.0, 7.0])
>>> preds = torch.tensor([2.5, 0.0, 2.0, 8.0])
>>> mean_absolute_error = MeanAbsoluteError()
>>> mean_absolute_error(preds, target)
tensor(0.5000)

MeanAbsolutePercentageError

MeanAbsolutePercentageError（MAPE）计算公式：

>>> from torchmetrics import MeanAbsolutePercentageError
>>> target = torch.tensor([1, 10, 1e6])
>>> preds = torch.tensor([0.9, 15, 1.2e6])
>>> mean_abs_percentage_error = MeanAbsolutePercentageError()
>>> mean_abs_percentage_error(preds, target)
tensor(0.2667)

MeanSquaredError

MeanSquaredError（MSE）计算公式：

Where is a tensor of target values, and is a tensor of predictions.

>>> from torchmetrics import MeanSquaredError
>>> target = torch.tensor([2.5, 5.0, 4.0, 8.0])
>>> preds = torch.tensor([3.0, 5.0, 2.5, 7.0])
>>> mean_squared_error = MeanSquaredError()
>>> mean_squared_error(preds, target)
tensor(0.8750)

MeanSquaredLogError

MeanSquaredLogError（MSLE）：

Where is a tensor of target values, and is a tensor of predictions.

>>> from torchmetrics import MeanSquaredLogError
>>> target = torch.tensor([2.5, 5, 4, 8])
>>> preds = torch.tensor([3, 5, 2.5, 7])
>>> mean_squared_log_error = MeanSquaredLogError()
>>> mean_squared_log_error(preds, target)
tensor(0.0397)

PearsonCorrcoef

PearsonCorrcoef计算公式：

Where is a tensor of target values, and is a tensor of predictions.

Forward accepts

• preds (float tensor): (N,)
• target``(float tensor): ``(N,)

>>> from torchmetrics import PearsonCorrcoef
>>> target = torch.tensor([3, -0.5, 2, 7])
>>> preds = torch.tensor([2.5, 0.0, 2, 8])
>>> pearson = PearsonCorrcoef()
>>> pearson(preds, target)
tensor(0.9849)

PSNR

自 v0.4 版后已弃用：PSNR 已移至 torchmetrics.image.psnr。 它将在 v0.5 中删除。

初始化内部模块状态，由 nn.Module 和 ScriptModule 共享。s

R2Score

R2Score，计算 r2 分数也称为决定系数：

![R^2 = 1 - \frac{SS_res}{SS_tot}](https://torchmetrics.readthedocs.io/en/stable/_images/math/05595643373c668b229056e8bf5c746aaafe40bd.png

其中，：残差平方和

：总平方和。

也可以计算调整后的 r2 分数由：

其中参数 k（独立回归量的数量）应作为调整后的参数提供。

Forward accepts

• preds (float tensor): (N,) or (N, M) (multioutput)
• target (float tensor): (N,) or (N, M) (multioutput)

在多输出的情况下，默认情况下，方差将在附加维度上均匀平均。 请参阅参数 multioutput 以更改此行为。

Example

>>>

>>> from torchmetrics import R2Score
>>> target = torch.tensor([3, -0.5, 2, 7])
>>> preds = torch.tensor([2.5, 0.0, 2, 8])
>>> r2score = R2Score()
>>> r2score(preds, target)
tensor(0.9486)

>>>

>>> target = torch.tensor([[0.5, 1], [-1, 1], [7, -6]])
>>> preds = torch.tensor([[0, 2], [-1, 2], [8, -5]])
>>> r2score = R2Score(num_outputs=2, multioutput='raw_values')
>>> r2score(preds, target)
tensor([0.9654, 0.9082])

SpearmanCorrcoef

计算 spearmans 等级相关系数。

其中 rg_x 和 rg_y 是与变量 x 和 y 关联的等级。 Spearmans 相关系数对应于对秩变量计算的标准 pearson 相关系数。

Example

>>>

>>> from torchmetrics import SpearmanCorrcoef
>>> target = torch.tensor([3, -0.5, 2, 7])
>>> preds = torch.tensor([2.5, 0.0, 2, 8])
>>> spearman = SpearmanCorrcoef()
>>> spearman(preds, target)
tensor(1.0000)

SSIM

自 v0.4 版后已弃用：SSIM 已移至 torchmetrics.image.ssim。 它将在 v0.5 中删除。

初始化内部模块状态，由 nn.Module 和 ScriptModule 共享。