deepextractor.utils.visualization

Module Contents

deepextractor.utils.visualization.save_predictions_as_plots(loader, model, folder='saved_predictions/', device='cuda')[source]

Save model prediction vs target plots for each sample in the loader.

deepextractor.utils.visualization.plot_examples(Difference_ts, clean_glitch_subtract, snrs, signal_type, PLOTS_PATH, indices_to_plot, noisy=False)[source]

Plot up to 3 example time series and save to disk.

deepextractor.utils.visualization.plot_q_transform(data, srate=4096.0, crop=None, whiten=True, ax=None, colourbar=True, qrange=[4, 64], frange=[10, 1200], clim=(0, 25.5))[source]

Plot the Q-transform of a time series using gwpy.

Parameters:

  • data (array-like) – Input time-domain data.

  • srate (float) – Sample rate in Hz.

  • crop (tuple or list, optional) – (center_time, duration) window in seconds for the Q-transform.

  • whiten (bool) – If True, apply whitening before the Q-transform.

  • ax (matplotlib.axes.Axes, optional) – Axes on which to plot. A new figure is created if not provided.

  • colourbar (bool) – If True, add a colorbar to the plot.

  • qrange (list) – [q_min, q_max] range for the Q-transform.

  • frange (list) – [f_min, f_max] frequency range in Hz.

  • clim (tuple) – (vmin, vmax) colour axis limits for normalised energy.