SINDy shows AttributeError: module 'numpy' has no attribute 'math' when attempting to fit a SINDy model.

[giopapanas]

Hi PySINDy community! I am writing a prototype PySINDy code for my data. I have 11 variables called: 'CVD60','t2m', 'ws', 'wd', 'tp', 'blh', 'ssr', 'no2', 'o3', 'so2', 'pm10'. I have these measurements taken everyday across several years, hence my time step is t=1 (1 day).

However, I keep on getting the error: AttributeError: module 'numpy' has no attribute 'math'. I also get this error when I was fitting the models in the PySINDY examples. I believe I have installed the environment and its dependencies fine.

Here is my prototype code (which I aim to optimize later, after I make it to work). Note that the X shape is (9127, 11) and the dt shape is (9127,):

feature_names = ['CVD60','t2m', 'ws', 'wd', 'tp', 'blh', 'ssr', 'no2', 'o3', 'so2', 'pm10']

Extract the features from the dataframe

X = data1[feature_names].values
dt = data1['Date'].values

print(X.shape)
print(dt.shape)

Define the differentiation method, feature library, and optimization algorithm

differentiation_method = ps.FiniteDifference(order=2)
feature_library = ps.PolynomialLibrary(degree=3)
optimizer = ps.STLSQ(threshold=0.2)

Create a SIMPLE SINDy model

model = ps.SINDy(differentiation_method=differentiation_method,
feature_library=feature_library,
optimizer=optimizer,feature_names=feature_names)

Fit the model to the data

model.fit(X, t=dt)
model.print()`

I am actually get the following error (entire copy):

---

AttributeError Traceback (most recent call last)
Cell In[41], line 36
31 model = ps.SINDy(differentiation_method=differentiation_method,
32 feature_library=feature_library,
33 optimizer=optimizer,feature_names=feature_names)
35 # Fit the model to the data
---> 36 model.fit(X, t=dt)
37 model.print()
39 # Loop through the dataframes and fit a SINDy model to each
40 #for i, data in enumerate(data_list):
41 # Features to use in the SINDy model
(...) 62 # opt = ps.LASSO(threshold=0.1, alpha=0.1)
63 # opt = ps.Ridge(alpha=0.1)

File c:\Users\giorg\pysindy\venv\Lib\site-packages\pysindy\pysindy.py:343, in SINDy.fit(self, x, t, x_dot, u, multiple_trajectories, unbias, quiet, ensemble, library_ensemble, replace, n_candidates_to_drop, n_subset, n_models, ensemble_aggregator)
337 u = validate_control_variables(
338 x,
339 u,
340 trim_last_point=(self.discrete_time and x_dot is None),
341 )
342 self.n_control_features_ = u[0].shape[u[0].ax_coord]
--> 343 x, x_dot = self._process_multiple_trajectories(x, t, x_dot)
345 # Set ensemble variables
346 self.ensemble = ensemble
...
425 return char.chararray
--> 427 raise AttributeError("module {!r} has no attribute "
428 "{!r}".format(name, attr))

AttributeError: module 'numpy' has no attribute 'math'

Do you believe it has to do with how PySINDY reads the multiple trajectories across numerous variables in X or smg else? Thank you very much in advance!