Shape error in analyse_cross_ff

[pipipyau]

I want to use analyse_cross_ff for my signal, but had error.

Steps to reproduce:

from StatTools.analysis.dfa import dfa
from StatTools.analysis.utils import analyse_cross_ff

s_vals, f2_vals = dfa(generated_signal, degree=2, processes=1, n_integral=1)
ff_params, _ = analyse_cross_ff(f2_vals, s_vals, crossover_amount=1)

My version is FluctuationAnalysisTools==1.9.0.

Error log:

ValueError                                Traceback (most recent call last)
Cell In[47], [line 2](vscode-notebook-cell:?execution_count=47&line=2)
      1 s_vals, f2_vals = dfa(generated_signal, degree=2, processes=1, n_integral=1)
----> [2](vscode-notebook-cell:?execution_count=47&line=2) ff_params, _ = analyse_cross_ff(f2_vals, s_vals, crossover_amount=1)

File ~/Documents/GitHub/StatTools/StatTools/analysis/utils.py:181, in analyse_cross_ff(hs, S, crossover_amount, max_ridigity, min_ridigity, min_slope_current, max_slope_current, ridigity_initial_parameter, slope_current_initial_parameter)
    171 p0 = (
    172     [
    173         0,
   (...)
    177     + [ridigity_initial_parameter] * r_count
    178 )
    180 # Perform an optimization
--> [181](https://file+.vscode-resource.vscode-cdn.net/Users/a1/Documents/GitHub/StatTools/research/kalman_filter/~/Documents/GitHub/StatTools/StatTools/analysis/utils.py:181) popt, pcov, infodict, mesg, ier = curve_fit(
    182     change_cross_value,
    183     np.log10(s.flatten()),
    184     np.log10(hs.flatten()),
    185     p0=p0,
    186     bounds=(
    187         bounds_min,
    188         bounds_max,
    189     ),
    190     full_output=True,
    191     maxfev=6000,
...
File ~/Documents/GitHub/StatTools/venv/lib/python3.13/site-packages/scipy/optimize/_minpack_py.py:[539](https://file+.vscode-resource.vscode-cdn.net/Users/a1/Documents/GitHub/StatTools/research/kalman_filter/~/Documents/GitHub/StatTools/venv/lib/python3.13/site-packages/scipy/optimize/_minpack_py.py:539), in _wrap_func.<locals>.func_wrapped(params)
    538 def func_wrapped(params):
--> 539     return func(xdata, *params) - ydata

ValueError: operands could not be broadcast together with shapes (64,) (8,)

s_vals.shape, f2_vals.shape print ((8,), (8,)). Maybe I'm doing something wrong?

[Sinitca-Aleksandr]

@pipipyau please, chech if the changes from #106 resolved your issue. If yes, close the issue.

[pipipyau]

If you use my code, the error remains.

from StatTools.analysis.dfa import dfa
from StatTools.analysis.utils import analyse_cross_ff

s_vals, f2_vals = dfa(generated_signal, degree=2, processes=1, n_integral=1)
ff_params, _ = analyse_cross_ff(f2_vals, s_vals, crossover_amount=1)

Error:

----> [2](vscode-notebook-cell:?execution_count=20&line=2) ff_params, _ = analyse_cross_ff(f2_vals, s_vals, crossover_amount=1)
ValueError: operands could not be broadcast together with shapes (121,) (11,) 

Shapes:
print(generated_signal.shape, f2_vals.shape) # (4096,) (11,)

Note:
If use np.expand_dims(f2_vals, axis=0) in analyse_cross_ff it works.

[Sinitca-Aleksandr]

I suppose, I makes sense to implement a feature to handle 1d input in analyse_cross_ff. @MayaGavrichenkova Could you manage it