vishack.core.evaluate¶
Evaluate statistical quantities from frequency series.
Functions
|
Maximum absolute error between the data and the reference |
|
Mean-square-error between the data and reference data. |
|
Calculated the expected root-mean-square value from the data |
|
Maximum absolute error between the whitened data and the reference |
|
Mean-square-error between the whitened data and reference data. |
|
Whiten the data and calcuate the expected root-mean-square value. |
-
vishack.core.evaluate.mae(data, reference)¶ Maximum absolute error between the data and the reference
- Parameters
data (array) – The data to be evaluted
reference (array) – The reference data.
- Returns
The maximum absolute error between the data and the reference
- Return type
float
-
vishack.core.evaluate.mse(data, reference)¶ Mean-square-error between the data and reference data.
- Parameters
data (array) – The data to be evaluted
reference (array) – The reference data.
- Returns
The mean-square-error between the data and the reference.
- Return type
float
-
vishack.core.evaluate.rms(data, df=1.0)¶ Calculated the expected root-mean-square value from the data
- Parameters
data (array) – The data to be evaluated
df (float, optional) – The frequency spacing between data points. Default to be 1.
- Returns
The expected RMS.
- Return type
float
Note
Complex arrays are accepted. Absolute values will be taken after whitening. If the data is an amplitude spectral density, then the output will be the expected RMS. If the data is a transfer function, then the output will be the 2-norm.
-
vishack.core.evaluate.wmae(data, reference)¶ Maximum absolute error between the whitened data and the reference
- Parameters
data (array) – The data to be evaluted
reference (array) – The reference data.
- Returns
The maximum absolute error between the data and the reference whitened by the inverse of the reference.
- Return type
float
-
vishack.core.evaluate.wmse(data, reference)¶ Mean-square-error between the whitened data and reference data.
- Parameters
data (array) – The data to be evaluted
reference (array) – The reference data.
- Returns
The mean-square-error between the data and the reference whitened by the inverse of the reference.
- Return type
float
-
vishack.core.evaluate.wrms(data, df=1.0, whitening=None)¶ Whiten the data and calcuate the expected root-mean-square value.
- Parameters
data (array) – The data to be evaluated.
df (float, optional) – The frequency spacing between data points. Default to be 1.
whitening (array, optional) – The whitening/weighting function. If None, default to be ones.
- Returns
The whitened expected RMS.
- Return type
float
Note
Complex arrays are accepted. Absolute values will be taken after whitening. If the data is an amplitude spectral density, then the output will be the expected RMS. If the data is a transfer function, then the output will be the 2-norm.