Skip to content

Latest commit

 

History

History
95 lines (82 loc) · 2.98 KB

CHANGELOG.md

File metadata and controls

95 lines (82 loc) · 2.98 KB
Raw

1.9

1.9.0

  • order by Neuropixel channels byshank / row / col: swapped the readers for spikeglx.Readers
  • output channel detection QC figure with thresholds
  • removed support for original channel detection
  • drop support for file list. This would have to be implemented outside with a reader object

1.8

1.8.0

  • Add the passingSpikes.table.pqt output
  • expose dredge and channel detection parameters

1.8.1

  • raise Runtime error if more than half of the channels are detected as bad
  • datashift patch when the number of spikes is insufficient to register the electrode signal

1.7

1.7.3

  • Update installation instructions for using cupy alongside torch
  • Remove some PyDantic and NumPy deprecations
  • Reworked integration test

1.7.1

  • bugfix geometry of probes in registration was hard coded to 384 channels
  • bugfix dredge detection memory management sets a number of jobs according to GPU size

1.6

1.6a00

1.6a.dev5

  • reworks datashift2 to use DARTsort spike detection, fed into DREDge motion estimation

1.6a.dev4

  • uses ibl-neuropixel 1.0.0 that contains the saturation mitigation in the pre-processing

1.6a.dev3

  • spike hole branch merge with overlap parameter set to 1024

1.6a.dev2

  • spike hole branch merge with overlap parameter set to 0

1.6a.dev1

  • adds APR RMS output to pre-processing

1.6a.dev0

  • outputs probe.channels.npy via phylib forced update

1.5

1.5.1 bugfix

  • probes.channel_labels was referenced in plural in some parts of the code

1.5.0

  • set minFR to 0.02 Hz to avoid too many near empty clusters

1.4

1.4.5

  • make pykilosort compatible with spikeglx 2023-04

1.4.4

  • remove annoying prints (PR#16)

1.4.1

  • bugfix: normalisation factor if whitening is used in pre-processing

1.4.0

  • stabilisation of the covariance matrix prior whitening computation
  • add QC outputs of the preprocessing phase

1.3

1.3.3

  • change integration tests infrastructure

1.3.2

  • support for NP2 geometries: label x coordinates of shanks at 200um, outputs shanks dataset

1.3.1

  • copy the _iblqc_ files to the original probe directory

1.3.0

  • Support for Neuropixel 2
  • read geometry from the spikeglx metadata file
  • GPU support for destriping

1.2

1.2.1

  • IBL pre-proc: add the channel removal code to the computing of the whitening matrix

1.2.0 alpha02

  • fix bug for the last template (Kush)

1.2.0 alpha01

  • ibllib > 2.5.0 pre-processing (destriping)
    • channel rejection and interpolation before destriping
    • uses the pykilosort parameters for the high-pass filter
    • multi-processing version of the destriping
  • QC:
    • destriping outputs the RMS of each batch after pre-processing
    • outputs

1.1

  • add pre-processing within pykilosort
  • whitening is optional and set as a parameter

1.0

1.0.1 2021-08-08

  • output the drift matrix in Alf format

1.0.2 2021-08-31

  • attempt to fix bugs introduced by chronic recordings that reduce amount of detected spikes