Skip to content

Latest commit

 

History

History
187 lines (151 loc) · 9.33 KB

CHANGELOG.md

File metadata and controls

187 lines (151 loc) · 9.33 KB
Raw

0.6.5

FIXES

  • CuPy deprecation fix (update methods to be compatible with latest CuPy)
  • imp -> importlib (Python deprecation)
  • Github Actions changes: use fully micromamba; actions v4; py 3.7 -> 3.10;

0.6.4

FEATURES

  • Torch inputs to cwt and ssq_cwt now supported

FEATURES (minor)

  • added complex=2 option to visuals.plot
  • added norm_scaling kwarg to visuals.imshow
  • extended visuals.hist
  • padsignal supports torch inputs

FIXES

  • _gmw.mom2cum: fix formula mistake, range(1, n - 1) -> range(1, n), by @bartvm
  • fix flipping ssq_freqs in torch
  • istft edge case with dtype='float32', time-localized window, and large hop_len
  • istft default N was supposed to be "largest possible" but wasn't; fixed
  • istft division by window norm now accounts for values below float tiny
  • icwt defaults args fix
  • fixed Ridge Extraction README
  • fixed important typo in ridge_extraction.py
  • fixed nv criterion for ignoring sanity check
  • fixed trigdiff for certain arg combinations
  • fix: librosa defaults to zero padding, which broke reconstruction_test

MISC

  • slight performance boost in _process_ssq_params
  • removed xp -= xp.mean(), a no-op with CWT, badly copied from MATLAB (original ssq code)
  • increased tolerance in infer_scaletype for float32, had false negatives
  • simplify utils.backend.astype
  • simplify p2up
  • nan_checks now defaults to True only for NumPy inputs
  • CI changes: conda -> micromamba
  • docstring references improved
  • docs: added missing references (w.r.t. original toolbox)
  • README section added
  • README examplesp improved
  • updated github workflows code

0.6.3 (1-23-2022): QoL, cleanups, fixes

FEATURES

  • freq_to_scale & scale_to_freq, experimental
  • Improved auto-scales
  • Optional nan_checks
  • Improved default gamma

BREAKING

  • ssq_freqs now correctly maps frequencies to corresponding rows of Tx for ssq_cwt, no longer requiring [::-1]
  • scales returned as 1D
  • extract_ridges: ridge_f (formerly fridge) now returns scales rather than log(scales)
  • extract_ridges: renamed: fridge -> ridge_f, max_energy -> ridge_e

FIXES

  • False warning for t & fs
  • Extended scope of astensor to scales, ssq_freqs
  • visuals: improve xticks & yticks handling

MISC

  • Added citation; README changes; docs changes
  • experimental: remove phase_squeeze, phase_transform
  • visuals.imshow: default cmap='turbo'
  • visuals: added auto_xlims, squeeze

0.6.1 (3-24-2021): GPU & CPU acceleration; caching

FEATURES

  • GPU acceleration & multi-thread CPU support for all forward transforms (cwt, stft, ssq_cwt, ssq_stft); see Performance guide
  • ssqueezepy.FFT, supporting single- & multi-threaded CPU execution, and GPU execution, optionally via pyfftw
  • dtype='float32' and 'float64' support for cwt, stft, ssq_cwt, ssq_stft, Wavelet
  • Wavelet.Psih(scale=, N=) will store the computed wavelet(s) and, if subsequent calls have identical scale and N, will return it directly without recomputing (significant speedup).

BREAKING

  • Dependency added: ftz
  • Default downsample: 3 -> 4 in utils.cwt_utils.make_scales
  • EPS deprecated in favor of EPS32 & EPS64 for respective precisions
  • ssq_cwt(flipud=True) default now returns Tx = np.flipud(Tx) relative to previous versions
  • ssq_cwt & ssq_stft number of variables returned now depend on get_w, get_dWx parameters; see docstrings
  • dtype defaults to 'float32' (can change via configs.ini); neither cwt nor stft, for most applications, require extreme precision like filters do, so defaults should prioritize compute
  • TestSignals.make_signals() now returns list of signals by default instead of dict with meta info (now accessible via get_params=True)

FIXES

  • ssq_stft would still default n_fft = len(x); defaulter line removed, delegated to stft.
  • ssqueezing: improperly handled return of infer_scaletype
  • ssqueezing: _get_center_frequency computed at N instead of p2up(N) with padtype != None

MISC

  • configs.ini: added new configurable defaults
  • ssqueezing.ssqueeze: added padtype arg (see FIXES)
  • ssqueezing.ssqueeze & ssq_cwt: added find_closest_parallel arg (see its docstring)
  • utils.cwt_utils: find_downsampling_scale added argument N
  • visuals.imshow: removed default 'interpolation' = 'none'
  • Added # Arguments: docstring to Wavelet
  • cwt significantly sped up: 1) per Wavelet reuse; 2) rid of ifftshift and *pn (they undo each other); 3) eliminated redundant allocation in vectorized

NOTES

  • Undocumented changes; skimming docstrings / source code should suffice for most purposes

0.6.0 (2-19-2021): Generalized Morse Wavelets, Ridge Extraction, Testing Suite

FEATURES (major)

  • Added Generalized Morse Wavelets (gmw, morsewave in _gmw.py)
  • Added automatic time-frequency ridge extraction, ridge_extraction.py
  • Added signal testing suite, _test_signals.py, and examples
  • Added higher-order CWT (via GMWs); _cwt.cwt_higher_order
  • Added configs.ini, used to control function defaults globally
  • Improved default scales to not over-represent low frequencies

FEATURES (other)

  • visuals: added wavelet_filterbank, viz_cwt_higher_order, viz_gmw_orders (first callable as wavelet.viz('filterbank'))
  • visuals.wavelet_tf: autopicks scale for scale=None to give a nice visual for any wavelet
  • ssq_cwt & ssq_stft: added arg preserve_transform to (see docstrings)
  • padsignal: 2D input support, of form (n_signals, signal_length) (i.e. will pad every row vector).
  • cwt: support for padtype=None
  • maprange: tuple of floats now supported (help(_ssq_cwt.ssq_cwt))
  • Wavelet.info() and @propertys of Wavelet revamped for generality; added @propertys: wc_ct, scalec_ct.
  • wavelets.center_frequency: added kind='peak-ct'
  • utils.find_max_scale now simpler and more effective, guaranteeing complete spectral coverage for low frequencies

BREAKING

  • utils.py -> utils/*: common.py, cwt_utils.py, stft_utils.py
  • The default wavelet has been changed from 'morlet' to 'gmw'
  • Changed Morlet's default parameters to closely match GMW's defaults per time & frequency resolution
  • ssq_cwt(mapkind=) default change: 'maximal' to 'peak'
  • scales default change: implicit preset from 'maximal' to 'minimal' for low scales, 'maximal' for high
  • ssq_cwt return order change: Tx, ssq_freqs, Wx, scales, w to Tx, Wx, ssq_freqs, scales, w, dWx (additionally returning dWx)
  • ssq_stft return order change: Tx, ssq_freqs, Sx, Sfs, dSx, w to Tx, Sx, ssq_freqs, Sfs, w, dSx
  • ssqueezing & ssq_cwt: renamed mapkind to maprange
  • difftype: 'direct' -> 'trig'
  • _infer_scaletype -> infer_scaletype
  • _integrate_analytic -> integrate_analytic
  • find_max_scale -> find_max_scale_alt, but find_max_scale is still (but a different) function

MISC

  • phase_cwt: takes abs(w) instead of zeroing negatives
  • wavelet in icwt and issq_cwt now defaults to the default wavelet
  • cwt: added args order, average
  • stft & ssq_stft: added t argument
  • stft default window increased frequency resolution
  • visuals.imshow(): cmap now defaults to 'jet' instead of 'bone' for abs=True
  • Added Examples to README, adjusted Minimal Example
  • NOTICE.txt: added jLab
  • setup.py: added short & long description, copyright, keywords

FIXES

  • visuals.wavelet_heatmap: string scales now functional
  • visuals: w overreached into negative frequencies for odd N in wavelet_tf, wavelet_tf_anim, & wavelet_heatmap
  • icwt: padtype now functional

FILE CHANGES

  • ssqueezepy/ added: _gmw.py, _test_signals.py, ridge_extraction.py, configs.ini, README.md
  • ssqueezepy/ added utils/, split utils.py into common.py, cwt_utils.py, stft_utils.py, __init__.py, & moved to utils/.
  • tests/ added: gmw_test.py, test_signals_test.py, ridge_extraction_test.py
  • examples/ added: extracting_ridges.py, scales_selection.py, ridge_extract_readme/: README.md, imgs/*
  • Created MANIFEST.in

0.5.5 (1-14-2021): STFT & Synchrosqueezed STFT

FEATURES

  • stft, istft, ssq_stft, and issq_stft implemented and validated
  • Added to utils.py: buffer, unbuffer, window_norm, window_resolution, and window_area
  • Replaced numba.njit with numba.jit(nopython=True, cache=True), accelerating recomputing

BREAKING

  • cwt() no longer returns x_mean
  • padsignal now only returns padded input by default; get_params=True for old behavior
  • Moved methods: phase_cwt & phase_cwt_num from ssqueezing to _ssq_cwt
  • In future release: return order of cwt and stft will be changed to have Wx, dWx and Sx, dSx, and ssq_cwt and ssq_stft to have Tx, Wx and Tx, Sx

MISC

  • wavelet positional argument in cwt is now a keyword argument that defaults to 'morlet'
  • Support for padsignal(padtype='wrap')
  • Docstring, comment cleanups