[Bug]: AttributeError: module 'scipy.signal' has no attribute 'cwt' (deprecated, use PyWavelets instead)

[tomvothecoder]

What happened?

Per https://docs.scipy.org/doc/scipy-1.12.0/reference/generated/scipy.signal.cwt.html:

    .. deprecated:: 1.12.0

        scipy.signal.cwt is deprecated in SciPy 1.12 and will be removed
        in SciPy 1.15. We recommend using PyWavelets instead.

What did you expect to happen? Are there are possible answers you came across?

• Add PyWavelets to dev.yml, ci.yml and pyproject.toml
• Replace scipy.signal.cwt with PyWavelets

  • e3sm_diags/e3sm_diags/driver/qbo_driver.py

    Lines 410 to 421 in df10271

    	Tuple(np.ndarray, np.ndarray)
    	The period and PSD arrays.
    	"""
    	deg = 6
    	period = np.arange(1, 55 + 1)
    	freq = 1 / period
    	widths = deg / (2 * np.pi * freq)
    	cwtmatr = scipy.signal.cwt(data, scipy.signal.morlet2, widths=widths, w=deg)
    	psd = np.mean(np.square(np.abs(cwtmatr)), axis=1)
    	return (period, psd)

• Update E3SM Unified notes that PyWavelets is a new dependency

Minimal Complete Verifiable Example (MVCE)

No response

Relevant log output

Traceback (most recent call last):
  File "/__w/e3sm_diags/e3sm_diags/e3sm_diags/parameter/core_parameter.py", line 340, in _run_diag
    single_result = module.run_diag(self)
  File "/__w/e3sm_diags/e3sm_diags/e3sm_diags/driver/qbo_driver.py", line 100, in run_diag
    test_dict["wave_period"], test_dict["wavelet"] = _calculate_wavelet(
  File "/__w/e3sm_diags/e3sm_diags/e3sm_diags/driver/qbo_driver.py", line 390, in _calculate_wavelet
    wave_period, wavelet = _get_psd_from_wavelet(detrended_data)
  File "/__w/e3sm_diags/e3sm_diags/e3sm_diags/driver/qbo_driver.py", line 418, in _get_psd_from_wavelet
    cwtmatr = scipy.signal.cwt(data, scipy.signal.morlet2, widths=widths, w=deg)
AttributeError: module 'scipy.signal' has no attribute 'cwt'. Did you mean: 'czt'?

Anything else we need to know?

Found in PR #907, where integration tests are breaking due to this issue.

Related SciPy docs:

• https://docs.scipy.org/doc/scipy-1.12.0/reference/generated/scipy.signal.morlet2.html

Related PyWavelets docs:

• https://pywavelets.readthedocs.io/en/latest/ref/cwt.html
• https://pywavelets.readthedocs.io/en/latest/ref/cwt.html#complex-morlet-wavelets

Environment

Latest e3sm_diags on main
scipy=1.15.1

[xylar]

@tomvothecoder, @chengzhuzhang, @whannah1, I would prefer not to delay Unified testing for this. Could we consider constraining scipy<1.15 in e3sm_diags for the time being while you all work this out?

A second question: do you all have a strategy for detecting deprecation warnings like this (e.g. as outputs from your testing) so they can be addressed before they become a last-minute problem? I know tests spit out a lot of deprecation warnings and sometimes they're related to other tools (like xarray) but they can sometimes be a helpful indication that something like this is coming.