metrics module

class metrics.AccuracyMetrics

Bases: metrics.Metrics

Accuracy metrics class.

name

The name of the metric.

Type

str

__init__()

Constuctor.

compute(y, scores)

Computes the accuracy of inferred numerical labels when compared to their true counterparts.

compute(y, scores)

Computes the accuracy of inferred numerical labels when compared to their true counterparts.

Parameters
  • y (numpy.ndarray) – True labels. Shape is (number of data points, )

  • scores (numpy.ndarray) – Activation of last layer of the model - the scores of the network. Shape is (batch_size, out_dim) where out_dim is the output dimension of the last layer of the model - usually same as the number of classes.

Returns

The accuracy of inferred numerical labels when compared to their true counterparts.

Return type

float

Notes

None

Raises

AssertionError – If y.shape is not the same as y_hat.shape

class metrics.MeanSquaredErrorMetrics

Bases: metrics.Metrics

MSE metrics class.

name

The name of the metric.

Type

str

__init__()

Constuctor.

compute(y, scores)

Computes the MSE of inferred numerical labels when compared to their true counterparts.

compute(y, scores)

Computes the MSE of inferred numerical labels when compared to their true counterparts.

Parameters
  • y (numpy.ndarray) – True labels. Shape is (number of data points, )

  • scores (numpy.ndarray) – Activation of last layer of the model - the scores of the network. Shape is (batch_size, out_dim) where out_dim is the output dimension of the last layer of the model - usually same as the number of classes.

Returns

The accuracy of inferred numerical labels when compared to their true counterparts.

Return type

float

Notes

None

Raises

AssertionError – If y.shape is not the same as y_hat.shape

class metrics.Metrics

Bases: object

Metrics parent class.

None
__init__()

Constuctor.