pyRheo is a Python package for rheological modeling, providing tools for creep, stress relaxation, small amplitude oscillatory shear, and steady shear flow models. This package is designed to help researchers and engineers analyze and model the behavior of viscoelastic materials.
Cite the peer-reviewed publication of pyRheo if you feel the package contributed to your research. Miranda-Valdez et al. Digital Discovery, 2025, DOI https://doi.org/10.1039/D5DD00021A
Feel free to share any suggestions and feedback.
Ask pyRheo's DeepWiki https://deepwiki.com/mirandi1/pyRheo
Version 1.0.0: First release of pyRheo
Version 1.0.1: OscillationModel and RotationModel classes were changed to SAOSModel and SteadyShearModel. Prompts will show up if a fractional order is close to 0 or 1
Version 1.0.2: Choose the cost function to minimize. For example:
model = RelaxationModel(model="FractionalZenerSolidS",
cost_function="BIC" # Other options are RSS, MAE, and MSE
initial_guesses="random",
num_initial_guesses=10,
minimization_algorithm="Powell",
mittag_leffler_type="Pade32"
)Refer to the docs/ folder to learn more about the package and how to utilize its API. The Documentation will be available via ReadtheDocs once the repository is public.
Install the pyRheo package directly from GitHub using pip:
pip install git+https://github.com/mirandi1/pyRheo.gitOnce the package has been installed, you can simply import its modules:
from pyRheo.creep_model import CreepModel
from pyRheo.relaxation_model import RelaxationModel
from pyRheo.oscillation_model import SAOSModel
from pyRheo.rotation_model import SteadyShearModelTo begin modeling data, first define a model object:
model = RelaxationModel(model="FractionalZenerSolidS", initial_guesses="random",
num_initial_guesses=10,
minimization_algorithm="Powell",
mittag_leffler_type="Pade32"
)fit the data with the model object:
model.fit(time, G_relax)Here, time and G_relax are Python NumPy arrays of the same size.
model.get_parameters()
model.print_parameters()
model.print_error()Plot the original data and the fitting results with
model.plot(time, G_relax)or use the fitted model to predict data over a customized time range
time_predict = np.logspace(np.min(np.log10(time)), np.max(np.log10(time)), 100)
G_relax_predict = model.predict(time_predict)Examples of the software's functionality can be found in the demos/ directory, which showcases how to use all the modules in pyRheo.
For a graphical user interface (GUI) of pyRheo, follow the instructions in the gui/ directory.
Inquiries and suggestions can be directed to isaac.mirandavaldez[at]aalto.fi or by raising an issue here.
The data used in the demos has been collected from:
K. Landauer, O. L. Kafka, N. H. Moser, I. Foster, B. Blaiszik and A. M. Forster, Scientific Data, 2023, 10, 356
F. A. Lavergne, P. Sollich and V. Trappe, The Journal of Chemical Physics, 2022, 156, 154901.
E. S. Epstein, L. Martinetti, R. H. Kollarigowda, O. Carey-De La Torre, J. S. Moore, R. H. Ewoldt and P. V. Braun, Journal of the American Chemical Society., 2019-02-27, 141, 3597– 3604.
I. Y. Miranda-Valdez, M. Sourroubille, T. Mäkinen, J. G. Puente-Córdova, A. Puisto, J. Koivisto and M. J. Alava, Cellulose, 2024, 31, 1545–1558.
R. G. Ricarte and S. Shanbhag, Polymer Chemistry, 2024, 15, 815–846.
K. R. Lennon, G. H. McKinley and J. W. Swan, Data-Centric Engineering, 2023, 4, e13.
R. I. Dekker, M. Dinkgreve, H. D. Cagny, D. J. Koeze, B. P. Tighe and D. Bonn, Journal of Non-Newtonian Fluid Mechanics, 2018, 261, 33–37.
V. A. H. Boudara, D. J. Read, J. Ramírez, Journal of Rheology, 2020, 64, 709-722.
Mittag-Leffler algorithms were implemented based on:
C. Zeng and Y. Q. Chen, Fractional Calculus and Applied Analysis, 2015, 18, 1492–1506.
I. O. Sarumi, K. M. Furati and A. Q. M. Khaliq, Journal of Scientific Computing, 2020, 82, 1–27.
R. Garrappa, SIAM Journal on Numerical Analysis, 2015, 53, 1350–1369.