List of changes in-between qpimage releases.

version 0.6.0

  • feat: automatically remove 2PI phase offsets when instantiating a QPImage (The phase offset is estimated from a 1px-wide border around the image)
  • feat: allow to disable the processing of phase data (unwrapping and correcting for phase offset) using proc_phase=False when instantiating a QPImage

version 0.5.4

  • docs: minor improvements

version 0.5.3

  • tests: minor improvements

version 0.5.2

  • enh: do not compress image data when HDF5 “core” driver is used
  • ref: fix deprecated .value (h5py)

version 0.5.1

  • ref: allow to subclass meta.MetaDict
  • docs: fixed several minor typos

version 0.5.0

  • feat: slicing of QPImage objects preserves background data, but background is merged from existing background data

version 0.4.6

  • docs: fix typos

version 0.4.5

  • docs: add HDF5 file format description

version 0.4.4

  • maintenance release

version 0.4.3

  • fix: use memory address in QPImage.__repr__ if identifier not given
  • fix: only use meta.DATA_KEYS in QPImage.__eq__
  • cleanup: remove unused “dm exclude” meta.DATA_KEYS definition

version 0.4.2

  • fix: minor identifier ambiguity

version 0.4.1

  • docs: add example of mask-based correction with qpsphere
  • docs: minor cleanup

version 0.4.0

  • BREAKING CHANGE: replace all occurences of “binary” with “mask” to avoid ambiguities

version 0.3.0

  • feat: new meta data key “sim model”

version 0.2.1

  • ci: automate PyPI release with travis-ci

version 0.2.0

  • drop support for Python 3.5
  • docs: add QPSeries example to user API section
  • feat: QPSeries.get_qpimage supports QPImage identifier as index
  • feat: allow to set data storage dtype, which now defaults to float32 to save disk space.
  • fix: set default gzip compression level to 9
  • ref: unify image data storage, set hdf5 chunks to image size

version 0.1.8

  • code cleanup

version 0.1.7

  • fix: bad default keyword argument in bg_estimate.estimate

version 0.1.6

  • implement qpimage refocusing function QPImage.refocus
  • API change: renamed “ramp” correction to “tilt” correction
  • add identifier to represenation string in QPImage
  • hologram analysis:
    • add disk, square, and tukey filters
    • standard “gauss” filter is replaced by “disk” filter
    • standard filter size is set to one third of the distance between the sideband and the central band
    • allow to set hologram-retrieval parameters as a keyword argument “holo_kw” in QPImage

version 0.1.5

  • support nan values in phase data
  • add fletcher32 checksums and gzip compression to hdf5 files (#10)
  • allow to hard-link background-correction data in QPSeries

version 0.1.4

  • more __init__ checks for QPImage and QPSeries (user convenience)
  • allow negative indices in QPSeries.get_qpimage
  • Bugfix: bg_estimate does not compute intersection but union (#9)

version 0.1.3

  • add QPImage.raw_amp and QPImage.raw_pha
  • improve QPImage.__eq__
  • add “identifier”:
    • meta data key
    • keyword for QPSeries.add_qpimage
    • property of and keyword for QPSeries
  • add convenience functions for item access in QPImage and QPSeries
  • moved to pathlib
  • minor API changes

version 0.1.2

  • allow strings and lists for which_data everywhere (#1)
  • check for valid background keys in (#2)
  • add property (#5)
  • add slicing (#6)
  • add references to documentation (#7)

version 0.1.1

  • QPImage.set_bg_data now accepts QPImage objects
  • add QPSeries for managing multiple QPImages in one hdf5 file (#3)

version 0.1.0

  • initial release