plot_utils module

Utility functions for plotting and preparing data for plots.

plot_utils.cov_pool(arr, axis=None, threshold=0.6, **kwargs)

Pooling function for downsampling correlation matrices for plotting, which can be passed as the func argument to skimage.measure.block_reduce.

The aim is to strike a balance between max pooling, which preserves important features (particularly the diagonal) but also artificially amplifies noise, and mean pooling which averages out noise but also washes out features. This function works by max pooling if the max is above some threshold value, and mean pooling otherwise.

Parameters

  • arr (ND numpy array) – Array to downsample.

  • axis (int, optional) – Axis to downsample, which is passed to np.max and np.mean. Default is None, which will cause those functions to use the flattened input.

  • threshold (float, optional) – Threshold for the max value in a pool, above which to use max pooling and below which to use mean pooling (default 0.6).

  • **kwargs – Additional keyword arguments to be passed to both np.max and np.mean.

plot_utils.get_3d_post(log_like_path, save_path)

Form 3D posterior grid from a log-likelihood file and save to disk, ready for plotting.

Parameters

  • log_like_path (str) – Path to log-likelihood text file.

  • save_path (str) – Path to save 3D posterior grid as .npz file.

plot_utils.get_cov_diags(cov_cng_fullsky_path, theory_cl_path, per_mask_data, lmin, lmax, lmax_mix, diags, save_path)

Extract diagonals of the covariance matrix for plotting with plotting.cov_diags.

Note that this is all for a single block, which in the paper is the auto-power in the lowest redshift bin.

Parameters

  • cov_cng_fullsky_path (str) – Path to the full-sky connected non-Gaussian covariance matrix.

  • theory_cl_path (str) – Path to theory power spectrum.

  • per_mask_data (list) – List of dictionaries, one dictionary per mask, each containing fields: mask_label used for the column headers, fsky sky fraction, mixmat_path path to mixing matrix or None for full sky, sim_cl_path path to simulated Cls as output by simulation.combine_sim_cl, cov_g_path path to Gaussian covariance, cov_ss_path path to super-sample covariance.

  • lmin (int) – Minimum l.

  • lmax (int) – Maximum l post-mixing.

  • lmax_mix (int) – Maximum l pre-mixing.

  • diags (list) – List of diagonals to extract, e.g. [0, 2, 10, 100].

  • save_path (str) – Path to save output data to.

plot_utils.get_cov_diags_gaussian(per_mask_data, diags, save_path)

Extract diagonals of the Gaussian covariance matrix for plotting with plotting.cov_gaussian.

Note that this is all for a single block, which in the paper is the auto-power in the lowest redshift bin.

Parameters

  • per_mask_data (list) – List of dictionaries, one dictionary per mask, each containing fields: mask_label used for the column headers, sim_cl_path path to simulated Cls as output by simulation.gaussian_sim, cov_g_path path to Gaussian covariance.

  • diags (list) – List of diagonals to extract, e.g. [0, 2, 10, 100].

  • save_path (str) – Path to save output data to.

plot_utils.get_cov_diags_withnoise(cov_cng_fullsky_path, per_mask_data, lmin, lmax, lmax_mix, diags, save_path)

Extract diagonals of the covariance matrix with noise (no simulations), for plotting with plotting.cov_withnoise.

Note that this is all for a single block, which in the paper is the auto-power in the lowest redshift bin.

Parameters

  • cov_cng_fullsky_path (str) – Path to the full-sky connected non-Gaussian covariance matrix.

  • per_mask_data (list) – List of dictionaries, one dictionary per mask, each containing fields: mask_label used for the column headers, fsky sky fraction, mixmat_path path to mixing matrix or None for full sky, cov_g_path path to Gaussian covariance, cov_ss_path path to super-sample covariance.

  • lmin (int) – Minimum l.

  • lmax (int) – Maximum l post-mixing.

  • lmax_mix (int) – Maximum l pre-mixing.

  • diags (list) – List of diagonals to extract, e.g. [0, 2, 10, 100].

  • save_path (str) – Path to save output data to.

plot_utils.get_cov_mats(cov_cng_fullsky_path, theory_cl_path, per_mask_data, lmin, lmax, lmax_mix, save_path)

Form correlation matrices for plotting with plotting.cov_mats.

Note that this is all for a single block, which in the paper is the auto-power in the lowest redshift bin.

Parameters

  • cov_cng_fullsky_path (str) – Path to the full-sky connected non-Gaussian covariance matrix.

  • theory_cl_path (str) – Path to theory power spectrum.

  • per_mask_data (list) – List of dictionaries, one dictionary per mask, each containing fields: mask_label used for the column headers, fsky sky fraction, mixmat_path path to mixing matrix or None for full sky, sim_cl_path path to simulated Cls as output by simulation.combine_sim_cl, cov_g_path path to Gaussian covariance, cov_ss_path path to super-sample covariance.

  • lmin (int) – Minimum l.

  • lmax (int) – Maximum l post-mixing.

  • lmax_mix (int) – Maximum l pre-mixing.

  • save_path (str) – Path to save output data to.