Source code for ACID_code_v2.acid

import warnings
warnings.filterwarnings("ignore")
import sys, emcee, os, time, inspect, inspect
import numpy as np
from math import log10, floor
from astropy.io import fits
from scipy.interpolate import interp1d
import multiprocessing as mp
from beartype import beartype
from numpy import integer as npint
import matplotlib.pyplot as plt
import scipy.constants as const
from . import utils
from .lsd import LSD
from . import mcmc
from .result import Result
from .data import Data

c_kms = float(const.c/1e3)

[docs] @beartype class Acid: """Accurate Continuum fItting and Deconvolution (ACID) class. This class contains the ACID method which fits the continuum of spectra and performs Least Squares Deconvolution (LSD) to obtain LSD profiles for each spectrum. It also contains many internal methods used within the main ACID function.""" def __init__( self, velocities :np.ndarray|None = None, linelist_path = None, linelist_wl :np.ndarray|list|None = None, linelist_depths:np.ndarray|list|None = None, verbose :int|npint|bool|None = 2, telluric_lines :np.ndarray|list|None = None, name :str = 'ACID', seed :int|npint|None = None, data = None, ): """Initialises the Acid class with inputted parameters. The parameters set here arre independent of the choice of the ACID and ACID_HARPS functions, which take different formats for inputted spectra. Parameters ---------- velocities : np.ndarray | None, optional Velocity grid for LSD profiles (in km/s). For example, use: np.arange(-25, 25, 0.82) to create. If None, a default grid from -25 to 25 km/s with a spacing calculated by calc_deltav. It is highly recommended to choose your own velocity grid, by default None linelist_path : str | None, optional Can be a path to linelist in string format, a dictionary with keys "wavelengths" and "depths", a Linelist class, or a list/array indexed such that 0 is the wavelengths and 1 is the depths. If None, you can directly provide linelist_wl and linelist_depths instead. At least one of linelist_path or linelist_wl and linelist_depths must be provided. By default None. linelist_wl : np.ndarray | list | None, optional Wavelengths of lines in linelist (in Angstroms). Only necessary if linelist_path is not provided. Must be same length as linelist_depths. If None, linelist_path must be provided., by default None linelist_depths : np.ndarray | list | None, optional Depths of lines in linelist (between 0 and 1). Only necessary if linelist_path is not provided. Must be same length as linelist_wl. If None, linelist_path must be provided., by default None verbose : bool | int | None, optional An integer between 0 and 3. If 0, nothing is printed. If 2, prints out useful progress information, as well as ACID warnings about any potential issues with the input data or autocorrelation warnings. If True, defaults to 2. If False, defaults to 0. If you want to ignore the warnings but still keep progress information, set verbose to 1. A verbosity of 3 will produce additional plots, such as the result of the continuum fit. By default None, which defaults to 2 (True). If set, overrides any verbose setting in the dataclass. telluric_lines : np.ndarray | list | None, optional List of wavelengths (in Angstroms) of telluric lines to be masked. This can also include problematic lines/features that should be masked also. For each wavelengths in the list ~3Å eith side of the line is masked. By default None name : str, optional Name to call any saved files, by default 'ACID' seed : int | None, optional Random seed for reproducibility, set it to None to be a random seed, by default 42 (the answer to life, the universe and everything) data : Data|None, optional An optional backend Data object to use for storing data. Allows previously calculated results to be skipped. If None, a new Data object is created, by default None. Please note that if the Data class already has a saved ACID config class, then those config values will overwrite the inputted values in initialisation or ACID method. """ # Initialise the data class to store calculations in ACID if data is not None: self.data = data else: self.data = Data() # Set config if old one exists self.config = self.data.config # Make config the same as old config, or generates a new empty one (handled in Data) # Validate velocities input, if None, this is handled in ACID function later when a input spectrum is provided if velocities is not None: if velocities.ndim != 1: raise ValueError("'velocities' must be a one-dimensional array") # data.velocities defaults to None in Data class, can be set in ACID function self.data.velocities = self.data.velocities if self.data.velocities is not None else velocities # Verbosity validation handled in config property setter self.config.verbose = verbose # Set linelist in the Data class, the property setter handles input validation self.data.set_linelist(linelist_path, linelist_wl, linelist_depths) self.config.telluric_lines = telluric_lines # Set seed if not already done in config, in this way, seed is only explicitly set once if getattr(self.config, "seed", None) is None: self.config.seed = seed if self.config.seed is not None: np.random.seed(self.config.seed) # In principle this is only ever called once # else: user may define a seed at the top of their seed, so can use that # else: seed already in config, so seed would already have been set when put in # I may make seed a property of the config class in the future # Name is also added to config self.config.name = name if getattr(self.config, "name", None) is None else self.config.name # Default order range for ACID, can be updated in ACID_HARPS. Eventually will add option to add this to inputs self.config.order_range = [1] # Save config to data class self.data.config = self.config # Determine if running in SLURM environment, independent of any previous configs self.slurm = "SLURM_JOB_ID" in os.environ self.sampler = None # sampler is a uniquely ACID attribute, so set here as needed in Results class return # Get init keys to be checked in ACID function for any potential conflicts in input arguments. # This is to avoid confusion for users who may accidentally input an argument that is meant for # the class initialisation rather than the ACID function, which takes different arguments. _INIT_KEYS = set(inspect.signature(__init__).parameters) - {"self"}
[docs] def ACID( self, wavelengths :list|np.ndarray|None = None, flux :list|np.ndarray|None = None, errors :list|np.ndarray|None = None, sn :int|np.ndarray|list|None = None, all_frames = None, deterministic_profile :bool = False, poly_ord :int|npint = 3, pix_chunk :int|npint = 20, dev_perc :int|npint = 25, n_sig :int|npint = 1, order :int|npint = 0, skips :int|npint = 1, parallel :bool = True, cores :int|npint|None = None, nsteps :int|npint = 10000, max_steps :int|npint|None = None, check_interval :int|npint = 1000, min_checks :int|npint = 1, min_tau_factor :int|npint = 50, tau_tol :float = 0.05, run_mcmc :bool = True, **kwargs, ): """Fits the continuum of the given spectra and performs LSD on the continuum corrected spectra, returning an LSD profile for each spectrum given. Spectra must cover a similiar wavelength range. Parameters ---------- wavelengths : np.ndarray | list | None, optional An array of wavelengths for each frame (in Angstroms). For multiple frames this should be a 2-d array such that wavelengths[i] corresponds to the wavelengths for the ith frame. Can only be None if a data instance was provided in initialisation. flux : np.ndarray | list | None, optional An array of spectral frames (in flux). For multiple frames this should be a 2-d array such that flux[i] corresponds to the spectral fluxes for the ith frame. Can only be None if a data instance was provided in initialisation., by default None errors : np.ndarray | list | None, optional Errors for each frame (in flux). For multiple frames this should be a 2-d array such that errors[i] corresponds to the spectral errors for the ith frame. Can only be None if a data instance was provided in initialisation., by default None sn : int | np.ndarray | list | None, optional Average signal-to-noise ratio for each frame (used to calculate minimum line depth to consider from line list). Each frame should have only one S/N value, so for multiple frames this should be a 1-d array such that sn[i] corresponds to the S/N for the ith frame. If None, the S/N will be estimated from the input spectra, by default None all_frames : str | np.ndarray | None, optional Output array for resulting profiles. Only neccessary if looping ACID function over many wavelength regions or order (in the case of echelle spectra). General shape needs to be (no. of frames, no. of orders, 2, no. of velocity pixels). If not provided, one is created with that shape. The only allowed string is "default" due to legacy behaviour, which now acts the same as None, by default None deterministic_profile : bool, optional If True, fits both the continuum and the LSD profile simultaneously. If False, only fits the continuum in mcmc, the profile is inferred from the continuum fit. Setting this to False can significantly speed up compution time, depending on the machine used as it is not as easy to parallelise. It may decrease accuracy, and is not fully tested as of yet, by default True. poly_ord : int, optional Order of polynomial to fit as the continuum, by default 3 pix_chunk : int, optional Size of 'bad' regions in pixels. 'bad' areas are identified by the residuals between an inital model and the data. If a residual deviates by a specified percentage (dev_perc) for a specified number of pixels, by default 20 dev_perc : int, optional Allowed deviation percentage. 'bad' areas are identified by the residuals between an inital model and the data. If a residual deviates by a specified percentage (dev_perc) for a specified number of pixels, by default 25 n_sig : int, optional Number of sigma to clip in sigma clipping. Ill fitting lines are identified by sigma-clipping the residuals between an inital model and the data. The regions that are clipped from the residuals will be masked in the spectra. This masking is only applied to find the continuum fit and is removed when LSD is applied to obtain the final profiles, by default 1 order : int, optional Only applicable if an all_frames output array has been provided as this is the order position in that array where the result should be input. i.e. if order = 5 the output profile and errors would be inserted in all_frames[:, 5]., by default 0 skips : int, optional An option to only select one in every n pixels, where n is the integer argument. This is only useful for testing to get a quick result, by default 1 parallel : bool, optional If True uses multiprocessing to calculate the profiles for each frame in parallel, by default True cores : int, optional Number of cores to use if parallel=True. If None, all available cores will be used, by default None nsteps : int, optional nsteps (int, optional): Number of steps for the MCMC to run, try increasing if it doesn't converge, by default 10000 max_steps : int | None, optional If set, the sampler will run until max_steps or convergence is reached by estimation using the emcee autocorrelation time (tau). The sampler will check for convergence every 'check_interval' steps, and will require a minimum number of checks ('min_checks') and a minimum tau factor ('min_tau_factor') before it can stop. The stopping criterion is met when the change in tau is less than 'tau_tol' for all parameters. By default None, which means no maximum. If a value is inputted, the nsteps parameter is ignored. The continue_sampling method in Result or Acid can still be used normally to continue after either stopping criterion is reached. check_interval : int, optional Interval (in steps) at which to check for MCMC convergence if max_steps is set, by default 1000. Only used if max_steps is set. min_checks : int, optional Minimum number of checks before MCMC can be stopped, by default 3. Only used if max_steps is set. min_tau_factor : int, optional Minimum tau factor for MCMC stopping criterion, by default 50. Only used if max_steps is set. tau_tol : float, optional Tolerance for tau convergence in MCMC stopping criterion, by default 0.01. Only used if max_steps is set. run_mcmc : bool, optional If True, runs the MCMC to fit the model, by default True. Can be set to False to perform all of the preparation for MCMC without actually running it. The ACID function will still update the class and data attributes. **kwargs : dict, optional Additional keyword arguments. kwargs are passed to the Result class when returning the Result object, see Result class for more details on what kwargs can be passed. Note that any kwargs that are also in the class initialisation will be ignored, and the inputted value will not be used. This is to avoid confusion for users who may accidentally input an argument that is meant for the class initialisation rather than the ACID function, which takes different arguments. The ignored kwargs are checked for and printed at the start of the function. Returns ------- Result Result object containing the LSD profiles and associated data. See Result class for methods and attributes. If run_mcmc is False, None is returned, but the class attributes are still updated (so that acid.data can be used for example). Raises ------ TypeError If the input types are not as expected. """ init_t0 = time.time() if self.config.verbose>1: print('Initialising...') # Setup and data validation done in data class and applies skips self.data.set_inputs(wavelengths, flux, errors, sn, skips) # Check for any potential conflicts in input arguments that are meant for the class initialisation. overlap = self._INIT_KEYS & kwargs.keys() if overlap and self.config.verbose > 0: for key in sorted(overlap): print(f"'{key}' is set in Acid initialisation, not the ACID method. The inputted value will be ignored.") # Raise an error if the kwargs are not part of either the ACID init or the Result init, so that the error happens # now rather than during the Result initialisation, which would waste the user's time valid_result_keys = set(inspect.signature(Result.__init__).parameters) - {"self"} invalid_keys = set(kwargs.keys()) - self._INIT_KEYS - valid_result_keys if invalid_keys and self.config.verbose > 0: raise ValueError(f"The following kwargs are not valid for either the ACID initialisation or the Result " f"initialisation: {', '.join(invalid_keys)}. Please check your input arguments.") # Assign inputted configuration to config dictionary plus or minus a few variables ACID_config = { "poly_ord" : poly_ord, "order" : order, "pix_chunk" : pix_chunk, "dev_perc" : dev_perc, "n_sig" : n_sig, "parallel" : parallel, "cores" : cores, "deterministic_profile" : deterministic_profile, "max_steps" : max_steps, "check_interval" : check_interval, "min_checks" : min_checks, "min_tau_factor" : min_tau_factor, "tau_tol" : tau_tol, "run_mcmc" : run_mcmc, } # TODO: make all input defaults None and overwrite config if input, with config handling problems caused therein # Update config if any of the above config settings are new self.config.update_lowpri(**ACID_config) # self.config overwrites ACID_config if overlapping self.data.config = self.config # update dataclass config as well if self.config.parallel and sys.platform == "win32": if self.config.verbose > 0: # This doesn't work, needs serious modifications to make work, so just run serially for now print("Parallel MCMC on Windows is not currently supported. Running MCMC serially.") self.config.parallel = False # Now that the data is set, we can check if the velocities were set in the initialisation or not, and if not, # calculate a default velocity grid using the input wavelengths. if self.data.velocities is None: if self.config.verbose > 0: print("Velocity grid not input, using a grid calculated from input wavelengths with default range of -25 to 25 km/s.\n " \ "It is highly recommended to input your own velocity grid, especially if you need a different wavelength range.") deltav = utils.calc_deltav(self.data.wavelengths["input"][0]) self.data.velocities = np.arange(-25, 25 + deltav, deltav) # default velocity grid from -25 to 25 km/s with spacing calculated from input wavelengths # Initiates all_frames variable, which is used to store the results of the MCMC sampling. # If an all_frames array is provided, this is used, otherwise a new one is created with the correct shape. self.data.initiate_all_frames(all_frames) ### Begin ACID process # Combines spectra from each frame (weighted based of S/N), returns to S/N of combined spectra. # If only one frame, just uses that frame (handled in the function). # This function requires assigned values: # self.data.wavelengths["input"], self.data.flux["input"], self.data.errors["input"], self.data.sn["input"] # To generate: # As of 1.0.4, this generates self.wavelengths["combined"], self.flux["combined"], self.errors["combined"] # As of 1.4.0, this now instead goes to the data class, so generates self.data.wavelengths["combined"], etc. # As of 1.4.0, this procedure is skipped if the outputs already exists in self.data to avoid recalculation if all(( hasattr(self.data.wavelengths, "combined"), hasattr(self.data.flux, "combined"), hasattr(self.data.errors, "combined"), hasattr(self.data.sn, "combined") )): if self.config.verbose > 2: print("Combined spectra already exists, skipping combination step.") else: if self.config.verbose > 2: print("Combining spectra...") self.combine_spec(output=False) # Clean combined spectra of NaNs wavelengths, flux, errors, nanmask = utils.drop_invalid(self.data.wavelengths["combined"], self.data.flux["combined"], self.data.errors["combined"], return_mask=True) self.data.wavelengths["combined"] = wavelengths self.data.flux["combined"] = flux self.data.errors["combined"] = errors self.data.nanmask = nanmask # Get the initial polynomial coefficents if not hasattr(self.data.wavelengths, "combined_normalized"): a, b = utils.get_normalisation_coeffs(self.data.wavelengths["combined"]) self.data.wavelengths["combined_normalized"] = (self.data.wavelengths["combined"]*a)+b # Compute an initial continuum fit # poly inputs has polynomial coefficients and scale at the end if all(( hasattr(self.data.flux, "fitted"), hasattr(self.data.errors, "fitted"), self.data.poly_inputs is not None )): if self.config.verbose > 2: print("Continuum fit already exists, skipping initial fit step.") else: if self.config.verbose > 2: print("Performing initial continuum fit...") self.data.poly_inputs, self.data.flux["fitted"], self.data.errors["fitted"] = self.continuumfit( self.data.flux["combined"], self.data.wavelengths["combined_normalized"], self.data.errors["combined"], poly_ord = self.config.poly_ord, ) self.data.wavelengths["fitted"] = np.copy(self.data.wavelengths["combined"]) # Just to keep track self.data.sn["fitted"] = np.copy(self.data.sn["combined"]) # SN also is not changed here if all(( self.data.model_inputs is not None, self.data.alpha is not None )): if self.config.verbose > 1: print("Initial LSD profile already exists, skipping initial LSD step.") else: if self.config.verbose > 1: print("Calculating initial LSD profile...") # Get the initial LSD profile using the initial fit initial_LSD = LSD(self.data) # Initialise LSD class with standard Acid attributes (verbosity, linelist, velocities, etc) initial_LSD.run_LSD(self.data.wavelengths["fitted"], self.data.flux["fitted"], self.data.errors["fitted"], self.data.sn["fitted"]) # Use alpha matrix and initial profile class variables from initial LSD run self.data.initial_profile = initial_LSD.profile # in optical depth self.data.initial_profile_errors = initial_LSD.profile_errors # Not used, saved for debugging self.data.alpha = initial_LSD.alpha # Set x, y, yerr, and model_inputs for emcee self.data.model_inputs = np.concatenate((self.data.initial_profile, self.data.poly_inputs)) # Masking based off residuals # Inputs: self.x, self.y, self.yerr, self.data.model_inputs, self.poly # Sets: self.c_factor, self.residual_masks # Modifies: self.alpha, self.yerr # As of 1.4.0, this is all applied to the data class (not internal ACID variables) if all(( self.data.residual_masks is not None, self.data.c_factor is not None )): if self.config.verbose > 1: print("Residual masks already exists, skipping residual masking step.") else: if self.config.verbose>1: print('Residual masking...') self.residual_mask() # will eventually add options for this ## Setting number of walkers and their start values(pos) self.data.ndim = len(self.data.model_inputs) factor = 3 # emcee recommendation self.data.nwalkers = self.data.ndim * factor rng = np.random.default_rng(self.config.seed) ### starting values of walkers with independent variation sigma = 0.8 * 0.005 initial_state = [] for i in range(0, self.data.ndim): if i < len(self.data.velocities): pos = rng.normal(self.data.model_inputs[i], sigma, (self.data.nwalkers, )) else: x1 = self.data.model_inputs[i] rounded_sigma = round(x1, 1-int(floor(log10(abs(x1))))-1) sigma = abs(rounded_sigma) / 10 pos = rng.normal(self.data.model_inputs[i], sigma, (self.data.nwalkers, )) initial_state.append(pos) if self.config.deterministic_profile is True: self.data.ndim = self.config.poly_ord + 1 self.data.nwalkers = self.data.ndim * factor initial_state = np.array(initial_state)[-self.data.ndim:, :self.data.nwalkers] # Transpose initial state to have shape (nwalkers, ndim) for emcee initial_state = np.transpose(np.array(initial_state)) self.data.initial_state = initial_state # Saved for debugging if needed, otherwise class variable not used for now ### ACID initialialised ### self.data.initialisation_time = time.time() - init_t0 if self.config.verbose>1: print('Initialised in %ss'%round((self.data.initialisation_time), 3)) if self.config.verbose>2: print('ACID Configuration before MCMC run:') print(f"Polynomial order: {self.config.poly_ord}") print(f"Deterministic profile: {self.config.deterministic_profile}") print(f"Number of walkers: {self.data.nwalkers}") print(f"Number of dimensions: {self.data.ndim}") # Run MCMC if self.config.run_mcmc is True: if self.config.max_steps is None: if self.config.verbose > 1: print("Running MCMC for %s steps..."%nsteps) self.run_mcmc(nsteps, initial_state) self.data.nsteps += nsteps else: if self.config.verbose > 1: print(f"Running MCMC with a maximum of {self.config.max_steps} steps or until convergence is reached...") self.run_mcmc_until_converged(self.config.max_steps, initial_state) self.data.nsteps = self.step_number self.data.mcmc_time = time.time() - init_t0 - self.data.initialisation_time return Result(self) else: if self.config.verbose > 0: print("MCMC not run, returning None. Class attributes have been updated.") return None
[docs] def ACID_HARPS( self, filelist : list, order_range : list|np.ndarray|None = None, save_path : str = './', file_type : str = 'e2ds', **kwargs, ): """ACID for HARPS e2ds and s1d spectra (DRS pipeline 3.5) Fits the continuum of the given spectra and performs LSD on the continuum corrected spectra, returning an LSD profile for each file given. Files must all be kept in the same folder as well as their corresponding blaze files. If 's1d' are being used their e2ds equivalents must also be in this folder. Result files containing profiles and associated errors for each order (or corresponding wavelength range in the case of 's1d' files) will be created and saved to a specified folder. It is recommended that this folder is seperate to the input files. Parameters ---------- filelist : list of strings List of files. Files must come from the same observation night as continuum is fit for a combined spectrum of all frames. A profile and associated errors will be produced for each file specified. order_range : array, optional Orders to be included in the final profiles. If s1d files are input, the corresponding wavelengths will be considered, by default None. save_path : str, optional Path to the directory where output files will be saved, by default './' file_type : str, optional Type of the input files, either "e2ds" or "s1d", by default 'e2ds' **kwargs Additional arguments to be passed to the ACID function. See ACID function for details. Returns ------- Object Result object containing the LSD profiles and associated data. ACID_HARPS=True flag is set to allow legacy subscripting and iteration if needed. The legacy subscript and iteration methods will access the following attributes: list Barycentric Julian Date for files list Profiles (in normalised flux) list Errors on profiles (in normalised flux) It can be accessed for example by: >>> result = Acid.ACID_HARPS(...) >>> BJDs = result.BJDs >>> profiles = result.profiles >>> errors = result.errors or >>> BJDs, profiles, errors = result """ file_type = file_type.lower() if file_type not in ['e2ds', 's1d']: raise ValueError("file_type must be either 'e2ds' or 's1d'") # Handle order_range input if order_range is None: # Be default, class is initialised with order_range = [1] for HARPS, this part forces # order range to np.arange(10, 70) if not specified for the ACID HARPS function. order_range = np.arange(10, 70) self.config.order_range = np.array(order_range) # Makes sure order range is an array regardless of input type self.file_type = file_type self.filelist = filelist for order in self.config.order_range: if self.config.verbose > 1: print('Running for order %s/%s...'%(order-min(self.config.order_range)+1, max(self.config.order_range)-min(self.config.order_range)+1)) frame_wavelengths, frame_flux, frame_errors, sns = self.read_in_frames(order, self.filelist, self.file_type) # Updates recursively the all_frames array with the profiles for each order self.ACID( frame_wavelengths, frame_flux, frame_errors, sns, order = order-min(self.config.order_range), **kwargs ) # adding into fits files for each frame BJDs = [] profiles = [] errors = [] for frame_no in range(0, len(frame_flux)): file = filelist[frame_no] fits_file = fits.open(file) hdu = fits.HDUList() hdr = fits.Header() for order in self.config.order_range: hdr['ORDER'] = order hdr['BJD'] = fits_file[0].header['ESO DRS BJD'] if order == self.config.order_range[0]: BJDs.append(fits_file[0].header['ESO DRS BJD']) hdr['CRVAL1'] = np.min(self.data.velocities) hdr['CDELT1'] = self.data.velocities[1] - self.data.velocities[0] profile = self.data.all_frames[frame_no, order-min(self.config.order_range), 0] profile_err = self.data.all_frames[frame_no, order-min(self.config.order_range), 1] hdu.append(fits.PrimaryHDU(data = [profile, profile_err], header = hdr)) if save_path != 'no save': month = 'August2007' hdu.writeto('%s%s_%s_%s.fits'%(save_path, month, frame_no, self.config.name), output_verify='fix', overwrite='True') result1, result2 = self.combineprofiles(self.data.all_frames[frame_no, :, 0], self.data.all_frames[frame_no, :, 1]) profiles.append(result1) errors.append(result2) self.BJDs = BJDs self.profiles = profiles self.errors = errors # Return Result class with ACID_HARPS=True flag to allow legacy subscripting and iteration if needed. return Result(self, ACID_HARPS=True)
[docs] def combine_spec( self, frame_wavelengths: np.ndarray | None = None, frame_flux: np.ndarray | None = None, frame_errors: np.ndarray | None = None, frame_sns: np.ndarray | None = None, output: bool = True ): """Combines multiple spectral frames into one spectrum Parameters ---------- frame_wavelengths : array, optional Wavelengths for the spectral frames, by default None frame_flux : array, optional Fluxes for the spectral frames, by default None frame_errors : array, optional Errors for the spectral frames, by default None frame_sns : array, optional Signal-to-noise ratio for the spectral frames, by default None output : bool, optional Whether to output the combined spectrum, by default True Returns ------- Tuple, if output is True, containing: combined_wavelengths : array Wavelengths for the combined spectrum combined_spectrum : array Fluxes for the combined spectrum combined_errors : array Errors for the combined spectrum combined_sn : float Signal-to-noise ratio for the combined spectrum """ if frame_wavelengths is not None: # This should only be for testing self.wavelengths["input"] = np.copy(frame_wavelengths) self.flux["input"] = np.copy(frame_flux) self.errors["input"] = np.copy(frame_errors) self.sn["input"] = np.copy(frame_sns) # Set simple names for variables (just used in this function) wavelengths = np.copy(self.data.wavelengths["input"]) flux = np.copy(self.data.flux["input"]) errors = np.copy(self.data.errors["input"]) sn = np.copy(self.data.sn["input"]) # Return as is if only one spectrum if len(self.data.wavelengths["input"])==1: self.data.wavelengths["combined"] = np.copy(self.data.wavelengths["input"][0]) self.data.flux["combined"] = np.copy(self.data.flux["input"][0]) self.data.errors["combined"] = np.copy(self.data.errors["input"][0]) self.data.sn["combined"] = np.copy(self.data.sn["input"][0]) else: # Get wavelength grid with highest S/N combined_wavelengths = wavelengths[np.argmax(sn)] interpolated_flux = np.zeros_like(flux) interpolated_errors = np.zeros_like(errors) # combine all spectra to one spectrum for n in range(len(flux)): # Interpolate each spectrum onto the combined wavelength grid f2 = interp1d(wavelengths[n], flux[n], kind = 'linear', bounds_error=False, fill_value = 'extrapolate') f2_err = interp1d(wavelengths[n], errors[n], kind = 'linear', bounds_error=False, fill_value = 'extrapolate') interpolated_flux[n] = f2(combined_wavelengths) interpolated_errors[n] = f2_err(combined_wavelengths) # Mask out out extrapolated areas idx_ex = np.logical_and(combined_wavelengths<=np.max(wavelengths[n]), combined_wavelengths>=np.min(wavelengths[n])) idx_ex = tuple([idx_ex==False]) interpolated_flux[n][idx_ex] = 1. interpolated_errors[n][idx_ex] = 1e12 # Mask out nans and zeros (these do not contribute to the main spectrum) where_are_NaNs = np.isnan(interpolated_flux[n]) interpolated_errors[n][where_are_NaNs] = 1e12 where_are_zeros = np.where(interpolated_flux[n] == 0)[0] interpolated_errors[n][where_are_zeros] = 1e12 where_are_NaNs = np.isnan(interpolated_errors[n]) interpolated_errors[n][where_are_NaNs] = 1e12 where_are_zeros = np.where(interpolated_errors[n] == 0)[0] interpolated_errors[n][where_are_zeros] = 1e12 invvars = 1 / interpolated_errors**2 invvars[interpolated_errors >= 1e12] = 0 weights = np.sum(invvars, axis=0) non_zero = weights > 0 weighted_flux = np.sum(interpolated_flux * invvars, axis=0) combined_flux = np.full_like(weights, 1.0) # or np.nan combined_errors = np.full_like(weights, 1e12) combined_flux[non_zero] = weighted_flux[non_zero] / weights[non_zero] combined_errors[non_zero] = 1 / np.sqrt(weights[non_zero]) frame_weights = np.sum(invvars, axis=1) combined_sn = np.sum(frame_weights * sn) / np.sum(frame_weights) self.data.wavelengths["combined"] = combined_wavelengths self.data.flux["combined"] = combined_flux self.data.errors["combined"] = combined_errors self.data.sn["combined"] = combined_sn if output is True: # ie if called as a function rather than from ACID function return combined_wavelengths, combined_flux, combined_errors, combined_sn
[docs] def continuumfit( self, fluxes : np.ndarray, wavelengths: np.ndarray, errors : np.ndarray, poly_ord : int|npint = 3, plot_result: bool = False ): """Provides an initial, normalised continuum fit using inputted spectra. Parameters ---------- fluxes : np.ndarray The flux values of the spectrum. wavelengths : np.ndarray The wavelengths corresponding to the spectrum. errors : np.ndarray The error values associated with the spectrum. poly_ord : int The order of the polynomial to fit to the continuum. By default 3. plot_result : bool, optional If True, plots the original spectrum and the fitted continuum, by default False Returns ------- tuple A tuple containing the polynomial coefficients, the normalized flux, and the normalized errors. """ m = np.isfinite(wavelengths) & np.isfinite(fluxes) w0 = wavelengths[m] f0 = fluxes[m] idx = np.argsort(w0) w = w0[idx] f = f0[idx] binsize = 100 n = len(w) // binsize # full bins only w2 = w[:n*binsize].reshape(n, binsize) f2 = f[:n*binsize].reshape(n, binsize) j = np.argmax(f2, axis=1) bins = np.arange(n) clipped_flux = f2[bins, j] clipped_waves = w2[bins, j] coeffs = np.polyfit(clipped_waves, clipped_flux, poly_ord) poly = np.poly1d(coeffs) fit = poly(wavelengths) flux_obs = fluxes / fit new_errors = errors / fit poly_coeffs = np.flip(coeffs) if self.config.verbose > 2 or plot_result is True: plt.figure(figsize=(10, 6)) plt.plot(wavelengths, fluxes, label='Original Spectrum') plt.plot(wavelengths, fit, label='Fitted Continuum', color='orange') plt.plot(clipped_waves, clipped_flux, 'o', label='Continuum Normalized Spectrum', color='green') plt.title('Continuum Fit') plt.legend() plt.show() return poly_coeffs, flux_obs, new_errors
[docs] def residual_mask( self, ): """Masks regions of the spectrum based on residuals from an initial model fit. """ ## iterative residual masking - mask continuous areas first - then possibly progress to masking the narrow lines # Set standard variables x = self.data.wavelengths["combined"] y = self.data.flux["combined"] yerr = self.data.errors["combined"] forward, _ = mcmc.MCMC(x, y, yerr, self.data.alpha).full_model(self.data.model_inputs) data_normalised = (y - np.min(y)) / (np.max(y) - np.min(y)) forward_normalised = (forward - np.min(forward)) / (np.max(forward) - np.min(forward)) residuals = data_normalised - forward_normalised ### finds consectuative sections where at least pix_chunk points have residuals greater than 0.25 - these are masked idx = (abs(residuals) > self.config.dev_perc / 100) flag_min = 0 flag_max = 0 for value in range(len(idx)): if idx[value] == True and flag_min <= value: flag_min = value flag_max = value elif idx[value] == True and flag_max < value: flag_max = value elif idx[value] == False and flag_max - flag_min >= self.config.pix_chunk: yerr[flag_min:flag_max] = 1e12 flag_min = value flag_max = value ############################################## # TELLURICS # ############################################## # self.yerr_compare = self.yerr.copy() ## masking tellurics for line in self.config.telluric_lines: limit = (21/c_kms)*line +3 idx = np.logical_and((line-limit) <= x, x <= (limit+line)) yerr[idx] = 1e12 # Note that this is used to keep track of the residual masks for later use in _get_profiles self.data.residual_masks = tuple([yerr >= 1e12]) ################################### ### sigma clip masking ### ################################### m = np.median(residuals) sigma = np.std(residuals) upper_clip = m + self.config.n_sig * sigma lower_clip = m - self.config.n_sig * sigma rcopy = residuals.copy() idx1 = tuple([rcopy <= lower_clip]) idx2 = tuple([rcopy >= upper_clip]) yerr[idx1] = 1e12 yerr[idx2] = 1e12 a, b = utils.get_normalisation_coeffs(x) poly_inputs, _bin, bye = self.continuumfit(y, (x*a)+b, yerr, self.config.poly_ord, plot_result=False) LSD_masking = LSD(self.data) LSD_masking.run_LSD(x, _bin, bye, sn=100) self.data.alpha = LSD_masking.alpha self.data.c_factor = LSD_masking.c_factor if self.config.verbose > 2: nremoved = np.sum(idx1)+np.sum(idx2) print(f"Residal masking has removed {nremoved}/{len(residuals)} points.") plt.figure(figsize=(10, 6)) plt.plot(x, residuals, label='Residuals', color='blue') plt.axhline(upper_clip, color='red', linestyle='--', label='Upper Clip Threshold') plt.axhline(lower_clip, color='green', linestyle='--', label='Lower Clip Threshold') plt.fill_between(x, upper_clip, lower_clip, color='gray', alpha=0.3, label='Clipped Region') for i, line in enumerate(self.config.telluric_lines): limit = (21/c_kms)*line + 3 label = 'Telluric Masking Region' if i==0 else None plt.axvspan(line-limit, line+limit, color='orange', alpha=0.5, label=label) plt.xlim(np.min(x), np.max(x)) plt.title('Residuals with Sigma Clipping Thresholds') plt.xlabel('Wavelength') plt.ylabel('Residuals') plt.legend(loc="lower right") plt.show() plt.figure(figsize=(10, 6)) plt.plot(self.data.velocities, LSD_masking.profile_F, label='LSD Profile after Masking and before sampling', color='red') plt.title('LSD Profile after Residual Masking') plt.xlabel('Velocity (km/s)') plt.ylabel('LSD Profile') plt.legend() plt.show() self.data.wavelengths["masked"] = x self.data.flux["masked"] = y # x and y dont change in this func self.data.errors["masked"] = yerr # yerr is modified in this func self.data.sn["masked"] = np.copy(self.data.sn["combined"]) # SN is not changed in this func # self.alpha is also modified in this func to get new alpha with masked residuals using pix chunk and dev perc return
[docs] def run_mcmc( self, nsteps, state = None, ): sampler_kwargs, mcmc_kwargs = self._get_sampler_kwargs(nsteps, state) if self.config.parallel: utils.configure_mp_environ(os) # Raises error is not configured correctly, otherwise does nothing if self.config.verbose>1: print(f"Using {self.config.cores} cores for MCMC") # For some reason, unspecified pooling as was before (as in case of windows in the else statement) # leds to a hung computer. So specify mp.get_context required, default is spawn, but spawn # causes multiple instances of this script to rerun, causing alpha matrix calculation to be redone # in each child process. Therefore, fork, which is legacy mp behavior on unix, is used. ctx = mp.get_context("fork") with ctx.Pool(processes=self.config.cores, initializer=mcmc._mp_init_worker, initargs=(self.data,)) as pool: self.sampler = emcee.EnsembleSampler(**sampler_kwargs, pool=pool, log_prob_fn=mcmc._mp_log_probability) self.sampler.run_mcmc(**mcmc_kwargs) else: MCMC = mcmc.MCMC(self.data) self.sampler = emcee.EnsembleSampler(**sampler_kwargs, log_prob_fn=MCMC) self.sampler.run_mcmc(**mcmc_kwargs)
[docs] def run_mcmc_until_converged( self, max_steps : int|npint, state=None, ): sampler_kwargs, mcmc_kwargs = self._get_sampler_kwargs(nsteps=self.config.check_interval, state=state) stopping_criterion_args = (self.config.min_checks, self.config.min_tau_factor, self.config.tau_tol) step_number = 0 tau_list = [] max_samples = max_steps // self.config.check_interval last_tolerance = np.inf last_neff = 0 if self.config.parallel: utils.configure_mp_environ(os) # Raises error is not configured correctly, otherwise does nothing ctx = mp.get_context("fork") with ctx.Pool(processes=self.config.cores, initializer=mcmc._mp_init_worker, initargs=(self.data,)) as pool: self.sampler = emcee.EnsembleSampler(**sampler_kwargs, pool=pool, log_prob_fn=mcmc._mp_log_probability) for i in range(max_samples): tol_str, neff_str = mcmc.MCMC.get_tqdm_desc(last_tolerance, last_neff, self.config) desc_dict = {"desc": f"Iteration {i+1}/{max_samples}, last tolerance: {tol_str}, neff: {neff_str}"} mcmc_kwargs["progress_kwargs"] = desc_dict self.sampler.run_mcmc(**mcmc_kwargs, skip_initial_state_check=True) mcmc_kwargs["initial_state"] = None # only use initial state for first run step_number += self.config.check_interval try: tau = self.sampler.get_autocorr_time(tol=0) except emcee.autocorr.AutocorrError: continue tau_list.append(tau) condition, last_tolerance, last_neff = mcmc.MCMC.get_mcmc_stopping_criterion(tau_list, step_number, *stopping_criterion_args) if condition is True and self.config.verbose > 1: print(f"Converged at step {step_number}. Final tolerance: {last_tolerance:.4f}, final effective sample size: {last_neff:.2f}.") break if self.config.verbose > 1 and condition is False: print(f"Not converged after reaching max steps of {step_number}. Final effective sample size: {last_neff:.2f}, final tolerance: {last_tolerance:.4f}.\n" f"Consider increasing max_steps.") else: MCMC = mcmc.MCMC(self.data) self.sampler = emcee.EnsembleSampler(**sampler_kwargs, log_prob_fn=MCMC) for i in range(max_samples): tol_str, neff_str = mcmc.MCMC.get_tqdm_desc(last_tolerance, last_neff, self.config) desc_dict = {"desc": f"Iteration {i+1}/{max_samples}, last tolerance: {tol_str}, neff: {neff_str}"} mcmc_kwargs["progress_kwargs"] = desc_dict self.sampler.run_mcmc(**mcmc_kwargs, skip_initial_state_check=True) mcmc_kwargs["initial_state"] = None # only use initial state for first run step_number += self.config.check_interval try: tau = self.sampler.get_autocorr_time(tol=0) except emcee.autocorr.AutocorrError: continue tau_list.append(tau) condition, last_tolerance, last_neff = MCMC.get_mcmc_stopping_criterion(tau_list, step_number, *stopping_criterion_args) if condition is True and self.config.verbose > 1: print(f"Converged at step {step_number}. Final tolerance: {last_tolerance:.4f}, final effective sample size: {last_neff:.2f}.") break if self.config.verbose > 1 and condition is False: print(f"Not converged after reaching max steps of {step_number}. Final effective sample size: {last_neff:.2f}, final tolerance: {last_tolerance:.4f}.\n" f"Consider increasing max_steps.") self.step_number = step_number
def _get_sampler_kwargs(self, nsteps, state=None): sampler_verbosity = True if self.config.verbose>1 else False backend = None if state is None: if not hasattr(self, 'sampler'): raise ValueError("No existing sampler found. Please run 'ACID' first or provide a state.") backend = self.sampler.backend # This includes previous seed if self.config.cores is None: if self.slurm: self.config.cores = int(os.environ.get("SLURM_CPUS_ON_NODE", 1)) else: self.config.cores = os.cpu_count() # TODO Make moves a ACID input moves = [ (emcee.moves.StretchMove(), 0.20), (emcee.moves.DESnookerMove(), 0.1), (emcee.moves.DEMove(), 0.6), (emcee.moves.DEMove(gamma0=1.0), 0.1) ] sampler_kwargs = { "nwalkers" : self.data.nwalkers, "ndim" : self.data.ndim, "moves" : moves, "backend" : backend, } mcmc_kwargs = { "initial_state": state, "nsteps" : nsteps, "progress" : sampler_verbosity, "store" : True, "tune" : True } return sampler_kwargs, mcmc_kwargs
[docs] def continue_sampling( self, sampler, nsteps : int|npint|None = None, max_steps : int|npint|None = None, max_steps_kwards : dict|None = None ): """Continue MCMC sampling for additional steps. This should be called in Result class by the user. This necessarily requires a Data instance to have been put into the ACID init. Parameters ---------- sampler : emcee.EnsembleSampler The existing MCMC sampler to continue sampling from. nsteps : int Number of additional steps to run the MCMC for. max_steps : int, optional Maximum number of steps to run the MCMC for in total (including previous steps). If specified, the MCMC will stop if this number of steps is reached even if convergence has not been reached, by default None. If input, nsteps is ignored. max_steps_kwards : dict, optional Additional keyword arguments to be passed to the run_mcmc_until_converged function if max_steps is specified, by default None. The kwargs description can be found in Acid.ACID(), they are the 4 kwargs appearing after max_steps. Typos for kwargs are silently ignored. Returns ------- emcee.EnsembleSampler The MCMC sampler after running for the additional steps. """ assert self.data.alpha is not None, "Data instance must have alpha matrix calculated to continue sampling." self.sampler = sampler self.config = self.data.config if max_steps is not None: if max_steps_kwards is not None: self.config.update_hipri(**max_steps_kwards) self.run_mcmc_until_converged(max_steps, state=None) # continue from current state self.data.nsteps += self.step_number else: self.run_mcmc(nsteps, state=None) # continue from current state self.data.nsteps += nsteps return self.sampler
[docs] def get_result( self=None, ): """Return a Result object for this instance or one passed explicitly. Parameters ---------- self : Acid instance, optional The Acid instance to get the Result for. If None, must be called on an instance of Acid. Returns ------- Result The Result object for the given Acid instance. """ if self is None: raise ValueError("Must be called on an instance or passed an instance explicitly") return Result(self)
[docs] def read_in_frames( self, order, filelist, file_type, directory=None, ): # read in first frame fluxes, wavelengths, flux_error_order, sn = LSD().blaze_correct( file_type, 'order', order, filelist[0], directory, 'unmasked', self.config.name, 'y') # fluxes, wavelengths, flux_error_order, sn, mid_wave_order, telluric_spec, overlap = LSD.blaze_correct( # file_type, 'order', order, filelist[0], directory, 'unmasked', self.config.name, 'y') frames = np.zeros((len(filelist), len(wavelengths))) errors = np.zeros((len(filelist), len(wavelengths))) frame_wavelengths = np.zeros((len(filelist), len(wavelengths))) sns = np.zeros((len(filelist), )) frames[0] = fluxes errors[0] = flux_error_order frame_wavelengths[0] = wavelengths sns[0] = sn def task_frames(frames, errors, frame_wavelengths, sns, i): file = filelist[i] frames[i], frame_wavelengths[i], errors[i], sns[i] = LSD().blaze_correct( file_type, 'order', order, file, directory, 'unmasked', self.config.name, 'y') return frames, frame_wavelengths, errors, sns ### reads in each frame and corrects for the blaze function, adds the spec, errors and sn to their subsequent lists for i in range(len(filelist[1:])+1): frames, frame_wavelengths, errors, sns = task_frames(frames, errors, frame_wavelengths, sns, i) ### finding highest S/N frame, saves this as reference frame idx = (sns==np.max(sns)) # global reference_wave reference_wave = frame_wavelengths[idx][0] reference_frame = frames[idx][0] reference_frame[reference_frame == 0] = 0.001 reference_error = errors[idx][0] reference_error[reference_frame == 0] = 1e12 # global frames_unadjusted frames_unadjusted = frames # global frame_errors_unadjusted frame_errors_unadjusted = errors ### each frame is divided by reference frame and then adjusted so that all spectra lie at the same continuum for n in range(len(frames)): f2 = interp1d(frame_wavelengths[n], frames[n], kind = 'linear', bounds_error=False, fill_value = 'extrapolate') div_frame = f2(reference_wave)/reference_frame idx_ref = (reference_frame<=0) div_frame[idx_ref]=1 binned = [] binned_waves = [] binsize = int(round(len(div_frame)/5, 1)) for i in range(0, len(div_frame), binsize): if i+binsize<len(reference_wave): waves = reference_wave[i:i+binsize] flux = div_frame[i:i+binsize] waves = waves[abs(flux-np.median(flux))<0.1] flux = flux[abs(flux-np.median(flux))<0.1] binned.append(np.median(flux)) binned_waves.append(np.median(waves)) binned = np.array(binned) binned_waves = np.array(binned_waves) ### fitting polynomial to div_frame try:coeffs = np.polyfit(binned_waves, binned, 4) except:coeffs = np.polyfit(binned_waves, binned, 2) poly = np.poly1d(coeffs) fit = poly(frame_wavelengths[n]) frames[n] = frames[n]/fit errors[n] = errors[n]/fit idx = (frames[n] == 0) frames[n][idx] = 0.00001 errors[n][idx] = 1e12 return frame_wavelengths, frames, errors, sns
[docs] def combineprofiles( self, spectra, errors, ): spectra = np.array(spectra) idx = np.isnan(spectra) shape_og = spectra.shape if len(spectra[idx])>0: spectra = spectra.reshape((len(spectra)*len(spectra[0]), )) for n in range(len(spectra)): if spectra[n] == np.nan: spectra[n] = (spectra[n+1]+spectra[n-1])/2 if spectra[n] == np.nan: spectra[n] = 0. spectra = spectra.reshape(shape_og) errors = np.array(errors) spectra_to_combine = [] weights=[] for n in range(0, len(spectra)): if np.sum(spectra[n])!=0: spectra_to_combine.append(list(spectra[n])) temp_err = np.array(errors[n, :]) weight = (1/temp_err**2) weights.append(np.mean(weight)) weights = np.array(weights/sum(weights)) spectra_to_combine = np.array(spectra_to_combine) length, width = np.shape(spectra_to_combine) spectrum = np.zeros((1,width)) spec_errors = np.zeros((1,width)) for n in range(0, width): temp_spec = spectra_to_combine[:, n] spectrum[0,n]=sum(weights*temp_spec)/sum(weights) spec_errors[0,n] = (np.std(temp_spec, ddof=1)**2) * np.sqrt(sum(weights**2)) spectrum = list(np.reshape(spectrum, (width,))) spec_errors = list(np.reshape(spec_errors, (width,))) return spectrum, spec_errors
[docs] def ACID(*args, **kwargs): """Legacy ACID function This function runs the legacy ACID code. This is provided for backwards compatibility with previous versions of ACID. It is recommended to use the ACID class and its methods for new code. The args and kwargs passing follows the original v1 version of ACID, which can be found in https://github.com/ldolan05/ACID Parameters ---------- *args Positional arguments to be passed to the ACID function. **kwargs Keyword arguments to be passed to the ACID initialisation and function. Returns ------- Any Returns the outputs of the ACID function (now a Result object). """ # Use old argument names and map to new ones LEGACY_ACID_ARGS = [ "input_wavelengths", "input_spectra", "input_spectral_errors", "line", "frame_sns", "vgrid", "all_frames", "poly_or", "pix_chunk", "dev_perc", "n_sig", "telluric_lines", "order", ] RENAMED_LEGACY_ARGS = { "input_wavelengths": "wavelengths", "input_spectra": "flux", "input_spectral_errors": "errors", "frame_sns": "sn", "vgrid": "velocities", "line": "linelist_path", "poly_or": "poly_ord", } # Split args and kwargs into init and run kwargs using helper function init_kwargs, run_kwargs = _get_init_and_run_kwargs(LEGACY_ACID_ARGS, RENAMED_LEGACY_ARGS, *args, **kwargs) acid = Acid(**init_kwargs) return acid.ACID(**run_kwargs)
[docs] def ACID_HARPS(*args, **kwargs): """Legacy ACID_HARPS function This function runs the legacy ACID_HARPS code. This is provided for backwards compatibility with previous versions of ACID. It is recommended to use the ACID class and its methods for new code. The args and kwargs passing follows the original v1 version of ACID_HARPS, which can be found in https://github.com/ldolan05/ACID Parameters ---------- *args Positional arguments to be passed to the run_ACID_HARPS function. **kwargs Keyword arguments to be passed to the ACID initialisation and run_ACID_HARPS function. Returns ------- Any Returns the outputs of the run_ACID_HARPS function (now a Result object). """ # Use old argument names and map to new ones LEGACY_HARPS_ARGS = [ "filelist", "line", "vgrid", "poly_or", "order_range", "save_path", "file_type", "pix_chunk", "dev_perc", "n_sig", "telluric_lines", ] RENAMED_LEGACY_ARGS = { "input_wavelengths": "wavelengths", "input_spectra": "flux", "input_spectral_errors": "errors", "frame_sns": "sn", "vgrid": "velocities", "line": "linelist_path", "poly_or": "poly_ord", } # Split args and kwargs into init and run kwargs using helper function init_kwargs, run_kwargs = _get_init_and_run_kwargs(LEGACY_HARPS_ARGS, RENAMED_LEGACY_ARGS, *args, **kwargs) acid = Acid(**init_kwargs) return acid.ACID_HARPS(**run_kwargs)
def _get_init_and_run_kwargs(legacy_args, renamed_args_map, *args, **kwargs): """Helper function to split legacy args and kwargs into init and run kwargs given legacy argument names and their renamed counterparts. """ legacy_kwargs = {} # Check for too many positional arguments if len(args) > len(legacy_args): raise TypeError(f"Too many positional arguments: {len(args)}") # Map positional arguments to their legacy names for i, val in enumerate(args): legacy_kwargs[legacy_args[i]] = val # Map legacy argument names to new ones translated_legacy = {} for key, val in legacy_kwargs.items(): new_key = renamed_args_map.get(key, key) translated_legacy[new_key] = val translated_kwargs = {} for key, val in kwargs.items(): new_key = renamed_args_map.get(key, key) translated_kwargs[new_key] = val # Combine both translated dictionaries combined = {**translated_legacy, **translated_kwargs} # Determine which arguments are for __init__ and which are for run_ACID_HARPS init_params = inspect.signature(Acid.__init__).parameters init_keys = set(init_params.keys()) - {"self"} # Split kwargs accordingly init_kwargs = {key: val for key, val in combined.items() if key in init_keys} run_kwargs = {key: val for key, val in combined.items() if key not in init_keys} return init_kwargs, run_kwargs