Skip to content

Latest commit

 

History

History
448 lines (301 loc) · 10.7 KB

plotting-tutorial.rst

File metadata and controls

448 lines (301 loc) · 10.7 KB
Raw

Plotting Tutorial

General Tips

Matplotlib Style Sheets

Set Style Sheet

import matplotlib.pyplot as plt
plt.style.use('seaborn-darkgrid')

Restore Default Style

import matplotlib
matplotlib.rcdefaults()

Quick Map Plot

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564
    ha.plot()

Quick Spectrum Plot

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Cube
    cube = Cube('8485-1901')
    spax = cube[17, 17]
    spax.flux.plot()

Quick Model Fit Plot

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')

    # must use Maps.getSpaxel() to get cube and modelcube
    spax = maps.getSpaxel(x=17, y=17, xyorig='lower', cube=True, modelcube=True)

    # mask out pixels lacking model fit
    no_fit = ~spax.full_fit.masked.mask

    # extra arguments to plot are passed to the matplotlib routine
    ax = spax.flux.plot(label='observed')
    ax.plot(spax.full_fit.wavelength[no_fit], spax.full_fit.value[no_fit], label='model')
    ax.legend()

Quick Image Plot

.. plot::
    :align: center
    :include-source: True

    import matplotlib.pyplot as plt
    from marvin.tools.image import Image
    image = Image(plateifu='8553-12702')
    image.plot()

BPT Plot

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    masks, fig, axes = maps.get_bpt()

Multi-panel Map Plot (Single Galaxy)

This code produces the right panel of Figure 1 from the Marvin paper.

.. plot::
    :align: center
    :include-source: True

    import matplotlib.pyplot as plt
    import numpy as np
    from marvin.tools import Maps

    maps = Maps('7977-12705')

    halpha = maps.emline_gflux_ha_6564
    nii_ha = np.log10(maps.emline_gflux_nii_6585 / halpha)
    stvel = maps.stellar_vel
    stsig = maps.stellar_sigma
    stsig_corr = stsig.inst_sigma_correction()

    with plt.style.context('seaborn-darkgrid'):
        fig, axes = plt.subplots(nrows=2, ncols=2, figsize=(12, 11))
        halpha.plot(fig=fig, ax=axes[0, 0])
        nii_ha.plot(fig=fig, ax=axes[0, 1], title="log([NII]6585 / H-alpha)", snr_min=None)
        stvel.plot(fig=fig, ax=axes[1, 0])
        stsig_corr.plot(fig=fig, ax=axes[1, 1])

Multi-panel Map Plot (Multiple Galaxies)

.. plot::
    :align: center
    :include-source: True

    import matplotlib.pyplot as plt
    from marvin.tools import Maps
    import marvin.utils.plot.map as mapplot

    plateifus = ['8485-1901', '7443-12701']
    mapnames = ['stellar_vel', 'stellar_sigma']

    with plt.style.context('seaborn-darkgrid'):
        rows = len(plateifus)
        cols = len(mapnames)
        fig, axes = plt.subplots(rows, cols, figsize=(8, 6))
        for row, plateifu in zip(axes, plateifus):
            maps = Maps(plateifu=plateifu)
            for ax, mapname in zip(row, mapnames):
                mapplot.plot(dapmap=maps[mapname], fig=fig, ax=ax, title=' '.join((plateifu, mapname)))

        fig.tight_layout()

Zoom-in Map Plot

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    fig, ax = ha.plot()
    ax.axis([13, 21, 13, 21])

Custom Map Colorbar Range Options

:align: center
:include-source: True

from marvin.tools import Maps
maps = Maps('8485-1901')
ha = maps.emline_gflux_ha_6564

fig, ax = ha.plot(percentile_clip=(1, 99))
fig, ax = ha.plot(sigma_clip=2)
fig, ax = ha.plot(cbrange=[2, 10])
fig, ax = ha.plot(symmetric=True)
fig, ax = ha.plot(log_cb=True)

Multi-panel Map Plot with Matching Colorbar Ranges

.. plot::
    :align: center
    :include-source: True

    import numpy as np
    import matplotlib.pyplot as plt
    from marvin.tools import Maps
    import marvin.utils.plot.map as mapplot

    maps = Maps('8485-1901')
    havel = maps.emline_gvel_ha_6564
    stvel = maps.stellar_vel
    vel_maps = [havel, stvel]

    cbranges = [vel_map.plot(return_cbrange=True) for vel_map in vel_maps]
    cb_max = np.max(np.abs(cbranges))
    cbrange = (-cb_max, cb_max)

    fig, axes = plt.subplots(ncols=2, figsize=(10, 4))
    for ax, vel_map in zip(axes, vel_maps):
        vel_map.plot(fig=fig, ax=ax, cbrange=cbrange)

    fig.tight_layout()

Custom Minimum Signal-to-Noise Ratio

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    # Default is 1 except for velocities, which default to 0.
    fig, ax = ha.plot(snr_min=10)

Custom No Usable IFU Data Region

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    # Defaults:
    # gray background (facecolor=''#A8A8A8'),
    # white lines (edgecolor='w'),
    # dense hatching: (hatch= 'xxxx')

    # Custom: black background, cyan lines, less dense hatching
    fig, ax = ha.plot(patch_kws={'facecolor': 'k',
                                 'edgecolor': 'c',
                                 'hatch': 'xx'})

Custom Axis and Colorbar Locations for Map Plot

.. plot::
    :align: center
    :include-source: True

    import matplotlib.pyplot as plt
    from marvin.tools import Maps

    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    fig = plt.figure()
    ax = fig.add_axes([0.12, 0.1, 2 / 3., 5 / 6.])
    fig, ax = ha.plot(fig=fig, ax=ax, cb_kws={'axloc': [0.8, 0.1, 0.03, 5 / 6.]})

Custom Spectrum and Model Fit

.. plot::
    :align: center
    :include-source: True

    import matplotlib.pyplot as plt
    from marvin.tools import Maps
    plt.style.use('seaborn-darkgrid')

    maps = Maps('1-209232')
    spax = maps.getSpaxel(x=0, y=0, xyorig='center', cube=True, modelcube=True)

    fig, ax = plt.subplots()

    pObs = ax.plot(spax.flux.wavelength, spax.flux.value)
    pModel = ax.plot(spax.full_fit.wavelength, spax.full_fit.value)
    pEmline = ax.plot(spax.emline_fit.wavelength, spax.emline_fit.value)
    plt.legend(pObs + pEmline + pModel, ['observed', 'emline model', 'model'])

    ax.axis([6700, 7100, -0.1, 3])
    ax.set_xlabel('observed wavelength [{}]'.format(spax.flux.wavelength.unit.to_string('latex')))
    ax.set_ylabel('flux [{}]'.format(spax.flux.unit.to_string('latex')))

Plot H\alpha Map of Star-forming Spaxels

.. plot::
    :align: center
    :include-source: True

    import numpy as np
    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564
    masks = maps.get_bpt(show_plot=False, return_figure=False)

    # Create a bitmask for non-star-forming spaxels by taking the
    # complement (`~`) of the BPT global star-forming mask (where True == star-forming)
    # and set bit 30 (DONOTUSE) for those spaxels.
    mask_non_sf = ~masks['sf']['global'] * ha.pixmask.labels_to_value('DONOTUSE')

    # Do a bitwise OR between DAP mask and non-star-forming mask.
    mask = ha.mask | mask_non_sf

    ha.plot(mask=mask)

Plot [NII]/H\alpha Flux Ratio Map of Star-forming Spaxels

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564
    nii = maps.emline_gflux_nii_6585
    nii_ha = nii / ha

    # Mask out non-star-forming spaxels
    masks, __, __ = maps.get_bpt(show_plot=False)

    # Create a bitmask for non-star-forming spaxels by taking the
    # complement (`~`) of the BPT global star-forming mask (where True == star-forming)
    # and set bit 30 (DONOTUSE) for those spaxels.
    mask_non_sf = ~masks['sf']['global'] * ha.pixmask.labels_to_value('DONOTUSE')

    # Do a bitwise OR between DAP mask and non-star-forming mask.
    mask = nii_ha.mask | mask_non_sf

    nii_ha.plot(mask=mask, cblabel='[NII]6585 / Halpha flux ratio')

Qualitative Colorbar

.. plot::
    :align: center
    :include-source: True

    import numpy as np
    from matplotlib import pyplot as plt
    from matplotlib.colors import ListedColormap
    from marvin.tools import Maps
    import marvin.utils.plot.map as mapplot

    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    # divide data into classes
    ha_class = np.ones(ha.shape, dtype=int)
    ha_class[np.where(ha.value > 5)] = 2
    ha_class[np.where(ha.value > 20)] = 3

    cmap = ListedColormap(['#104e8b', '#5783ad', '#9fb8d0'])
    fig, ax, cb = mapplot.plot(dapmap=ha, value=ha_class, cmap=cmap, cbrange=(0.5, 3.5),
                               title='', cblabel='Class', return_cb=True)
    cb.set_ticks([1, 2, 3])
    cb.set_ticklabels(['I', 'II', 'III'])

Custom Values and Custom Mask

.. plot::
    :align: center
    :include-source: True

    from marvin.tools import Maps
    import marvin.utils.plot.map as mapplot

    maps = Maps('8485-1901')
    ha = maps.emline_gflux_ha_6564

    # Mask spaxels without IFU coverage
    # nocov = ha.mask & 2**0
    nocov = ha.pixmask.get_mask('NOCOV')

    # Mask spaxels with low Halpha flux
    low_ha = (ha.value < 6) * ha.pixmask.labels_to_value('DONOTUSE')

    # Combine masks using bitwise OR (`|`)
    mask = nocov | low_ha

    fig, ax = mapplot.plot(dapmap=ha, value=ha.value, mask=mask)