Lognormal 2mode

Particula-beta Index / Particula Beta / Data / Process / Lognormal 2mode

Auto-generated documentation for particula_beta.data.process.lognormal_2mode module.

Attributes

• logger - Set up logging: logging.getLogger('particula')

cost_function

Show source in lognormal_2mode.py:28

Cost function for the lognormal distribution with 2 modes.

Arguments

• params - Combined array of mode_values, geometric_standard_deviation, and number_of_particles.
• x_values - The x-values (particle sizes).
• concentration_pdf - The actual concentration PDF to fit.

Returns

The mean squared error between the actual and guessed concentration PDF.

Signature

def cost_function(
    params: NDArray[np.float64],
    x_values: NDArray[np.float64],
    concentration_pdf: NDArray[np.float64],
) -> float: ...

create_lognormal_2mode_from_fit

Show source in lognormal_2mode.py:385

Create a fitted PMF stream and concentration matrix based on optimized parameters.

Arguments

• parameters_stream - Stream object containing the optimized parameters.
• radius_min - Log10 of the minimum radius value in meters (default: -9).
• radius_max - Log10 of the maximum radius value in meters (default: -6).
• num_radius_bins - Number of radius bins to create between radius_min and radius_max.

Returns

• fitted_pmf_stream - A Stream object containing the time and fitted concentration PMF data.
• fitted_concentration_pmf - A numpy array with the fitted concentration PMF values.

Signature

def create_lognormal_2mode_from_fit(
    parameters_stream: Stream,
    radius_min: float = 1e-09,
    radius_max: float = 1e-06,
    num_radius_bins: int = 250,
) -> Tuple[Stream, NDArray[np.float64]]: ...

See also

• Stream

evaluate_fit

Show source in lognormal_2mode.py:133

Evaluate the best fit and calculate R² score.

Signature

def evaluate_fit(
    best_result: dict[str, Any],
    logspace_x: NDArray[np.float64],
    concentration_pdf: NDArray[np.float64],
) -> Tuple[NDArray[np.float64], NDArray[np.float64], NDArray[np.float64], float]: ...

get_bounds

Show source in lognormal_2mode.py:75

Provide default bounds for optimization parameters.

Signature

def get_bounds() -> List[Tuple[float, Any]]: ...

get_initial_guesses

Show source in lognormal_2mode.py:87

Combine initial guesses into a single array.

Signature

def get_initial_guesses(
    mode_guess: NDArray[np.float64],
    geometric_standard_deviation_guess: NDArray[np.float64],
    number_of_particles_in_mode_guess: NDArray[np.float64],
) -> NDArray[np.float64]: ...

guess_and_optimize_looped

Show source in lognormal_2mode.py:303

Generate initial guesses using a machine learning model, optimize them, and return a Stream object with the results.

Arguments

• experiment_time - Array of experiment time points.
• radius_m - Array of particle radii in meters.
• concentration_m3_pdf - 2D array of concentration PDFs for each time point.

Returns

• fitted_stream - A Stream object containing the initial guesses, optimized values, and R² scores.

Signature

def guess_and_optimize_looped(
    experiment_time: NDArray[np.float64],
    radius_m: NDArray[np.float64],
    concentration_m3_pdf: NDArray[np.float64],
) -> Stream: ...

See also

• Stream

optimize_fit

Show source in lognormal_2mode.py:167

Optimize the lognormal 2-mode distribution parameters using multiple optimization methods.

Signature

def optimize_fit(
    mode_guess: NDArray[np.float64],
    geometric_standard_deviation_guess: NDArray[np.float64],
    number_of_particles_in_mode_guess: NDArray[np.float64],
    logspace_x: NDArray[np.float64],
    concentration_pdf: NDArray[np.float64],
    bounds: Optional[List[Tuple[float, Any]]] = None,
    list_of_methods: Optional[List[str]] = None,
) -> Tuple[
    NDArray[np.float64], NDArray[np.float64], NDArray[np.float64], float, dict[str, Any]
]: ...

optimize_fit_looped

Show source in lognormal_2mode.py:234

Loop through the concentration PDFs to get the best optimization.

Arguments

• mode_guess - Array of mode values.
• geometric_standard_deviation_guess - Array of geometric standard deviations.
• number_of_particles_in_mode_guess - Array of number of particles.
• x_values - Array of x-values (particle sizes).
• concentration_pdf - Matrix of concentration PDF values.
• bounds - List of bounds for optimization.
• list_of_methods - List of optimization methods.

Returns

• optimized_mode_values - Optimized mode values.
• optimized_gsd - Optimized geometric standard deviations.
• optimized_number_of_particles - Optimized number of particles.
• r2 - R² score.
• optimization_results - Dictionary of optimization results.

Signature

def optimize_fit_looped(
    mode_guess: NDArray[np.float64],
    geometric_standard_deviation_guess: NDArray[np.float64],
    number_of_particles_in_mode_guess: NDArray[np.float64],
    logspace_x: NDArray[np.float64],
    concentration_pdf: NDArray[np.float64],
    bounds: Optional[List[Tuple[float, Any]]] = None,
    list_of_methods: Optional[List[str]] = None,
) -> Tuple[
    NDArray[np.float64], NDArray[np.float64], NDArray[np.float64], NDArray[np.float64]
]: ...

run_optimization

Show source in lognormal_2mode.py:102

Perform the optimization using the specified method.

Signature

def run_optimization(
    method: str,
    initial_guess: NDArray[np.float64],
    bounds: List[Tuple[float, Any]],
    x_values: NDArray[np.float64],
    concentration_pdf: NDArray[np.float64],
) -> Optional[dict[str, Any]]: ...