skferm.datasets package#
Submodules#
skferm.datasets.mtp_ph module#
- files(anchor: ModuleType | str | None = None) Traversable[source]#
Get a Traversable resource for an anchor.
skferm.datasets.rheolaser module#
- files(anchor: ModuleType | str | None = None) Traversable[source]#
Get a Traversable resource for an anchor.
- clean_rheolaser(df: DataFrame, cutoff: int | None = None) DataFrame[source]#
Function which transforms the raw rheolaser format to a nice long format.
- load_rheolaser_data(clean: bool = True, cutoff: int | None = None) DataFrame[source]#
Load the Rheolaser dataset from the package resources. This is the exact format you get from a Rheolaser machine. Use clean_rheolaser to transform it into a long format DataFrame.
- Returns:
DataFrame containing the Rheolaser dataset.
- Return type:
pd.DataFrame
skferm.datasets.synthetic module#
- gompertz(t: Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | complex | bytes | str | _NestedSequence[complex | bytes | str], a: float, b: float, c: float)[source]#
Predict the value at time t using the modified Gompertz function.
Parameters: t (float or array-like): The time at which to predict the value. a (float): The upper asymptote. b (float): The displacement along the time axis. c (float): The growth rate.
Returns: float or array-like: The function value at time t.
- modified_gompertz(t: Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | complex | bytes | str | _NestedSequence[complex | bytes | str], A: float, L: float, mu: float)[source]#
modified gompertz as proposed by Zwietering et al. 1990
This gompertz has more interpretable parameters than the original gompertz.
- Parameters:
Returns: float or array-like: The function value at time t.
- logistic_growth(t: Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | complex | bytes | str | _NestedSequence[complex | bytes | str], N0: float, Nmax: float, r: float) ndarray[source]#
Simulate microbial growth using the logistic growth model.
Parameters: - t (array-like): Time points. - N0 (float): Initial population size. - Nmax (float): Maximum population size (carrying capacity). - r (float): Growth rate.
Returns: - array-like: Population at each time point.
- generate_synthetic_growth(time: Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | complex | bytes | str | _NestedSequence[complex | bytes | str], model: str = 'logistic', noise_std: float = 0.0, **kwargs) DataFrame[source]#
Generate synthetic growth data using specified growth model.
- Parameters:
(array-like) (- time)
(str) (- model)
(float) (- noise_std)
**kwargs (Parameters for the growth model.)
- Returns:
- dict
- Return type:
A dictionary with time and population arrays.
Module contents#
Dataset utilities for scikit-ferm.
- generate_synthetic_growth(time: Buffer | _SupportsArray[dtype[Any]] | _NestedSequence[_SupportsArray[dtype[Any]]] | complex | bytes | str | _NestedSequence[complex | bytes | str], model: str = 'logistic', noise_std: float = 0.0, **kwargs) DataFrame[source]#
Generate synthetic growth data using specified growth model.
- Parameters:
(array-like) (- time)
(str) (- model)
(float) (- noise_std)
**kwargs (Parameters for the growth model.)
- Returns:
- dict
- Return type:
A dictionary with time and population arrays.
- load_rheolaser_data(clean: bool = True, cutoff: int | None = None) DataFrame[source]#
Load the Rheolaser dataset from the package resources. This is the exact format you get from a Rheolaser machine. Use clean_rheolaser to transform it into a long format DataFrame.
- Returns:
DataFrame containing the Rheolaser dataset.
- Return type:
pd.DataFrame