Merge pull request #17 from LRydin/v0.4

Merging with dev.

.travis.yml

@@ -1,10 +1,17 @@
 language: python
 python:
-#  - "3.4"
-#  - "3.5"
+  - "3.4"
+  - "3.5"
   - "3.6"
   - "3.7"
-#  - "nightly"
+  - "3.8"
+  - "3.9"
+  - "nightly"
+
+arch:
+  - amd64
+  - arm64
+
 # command to install dependencies
 
 before_install:
@@ -14,9 +21,25 @@ install:
   - pip install coverage
   - pip install -r requirements.txt
 
-# command to run tests
+jobs:
+  include:
+    - stage: "Extra packages"
+      name: "Testing extra packages"
+      python: 3.6
+      install:
+        - pip install coverage
+        - pip install -r requirements.txt
+        - pip install EMD-signal
+      script:
+        - coverage run -m pytest test/test_EMD.py
+        - coverage run -m pytest test/test_MFDFA_extras.py
+  allow_failures:
+    - python: "nightly"
+
+
 script:
-  coverage run -m pytest test/
+  - coverage run -m pytest test/test_fgn.py
+  - coverage run -m pytest test/test_MFDFA.py
 
 after_success:
   bash <(curl -s https://codecov.io/bash)

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2019 Rydin
+Copyright (c) 2019-2021 Rydin
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

MFDFA/emddetrender.py

@@ -2,9 +2,11 @@
 # Empirical Mode Decomposition algorithm', https://github.com/laszukdawid/PyEMD,
 # licenced under the Apache 2.0 Licencing.
 
-from PyEMD import EMD
+# from PyEMD import EMD
 import numpy as np
 
+# Import of PyEMD is called inside the function
+
 def detrendedtimeseries(timeseries: np.ndarray, modes: list):
     """
     The function calculates the Intrinsic Mode Functions (IMFs) of a given
@@ -50,6 +52,7 @@ def detrendedtimeseries(timeseries: np.ndarray, modes: list):
 
     return detrendedTimeseries
 
+
 def IMFs(timeseries: np.ndarray):
     """
     Extract the Intrinsic Mode Functions (IMFs) of a given timeseries.
@@ -62,14 +65,19 @@ def IMFs(timeseries: np.ndarray):
     Notes
     -----
     .. versionadded:: 0.3
-    
+
     Returns
     -------
     IMFs: np.ndarray
         The Intrinsic Mode Functions (IMFs) of the Empirical Mode Decomposition.
         These are of shape ``(..., timeseries.size)``, with the first dimension
         varying depending on the data. Last entry is the residuals.
     """
+
+    # Check if EMD-signal is installed
+    missing_library()
+    from PyEMD import EMD
+
     # Initiate pyEMD's EMD function
     emd = EMD()
 
@@ -78,3 +86,13 @@ def IMFs(timeseries: np.ndarray):
 
     # Returns the IMFs as a (..., timeseries.size) numpy array.
     return IMFs
+
+
+def missing_library():
+    try:
+        import PyEMD.EMD as _EMD
+    except ImportError:
+        raise ImportError(("PyEMD is required to do Empirical Mode "
+                           "decomposition. Please install PyEMD with 'pip "
+                           "install EMD-signal'.")
+                         )

README.md

@@ -107,6 +107,7 @@ Let `Xₜ` be a multi fractal stochastic process. This mean `Xₜ` scales with s
 With the help of taking different powers variations of the `DFA`, one we can distinguish monofractal and multifractal processes.
 
 # Changelog
+- Version 0.4 - EMD is now optional. Restored back compatibility: py3.3 to py3.9
 - Version 0.3 - Adding EMD detrending. First release. PyPI code.
 - Version 0.2 - Removed experimental features. Added documentation
 - Version 0.1 - Uploaded initial working code
@@ -125,7 +126,7 @@ This project was started in 2019 at the [Faculty of Mathematics, University of O
 
 
 ### Funding
-Helmholtz Association Initiative *Energy System 2050 - A Contribution of the Research Field Energy* and the grant No. VH-NG-1025, *STORM - Stochastics for Time-Space Risk Models* project of the Research Council of Norway (RCN) No. 274410, and the *E-ON Stipendienfonds*.
+Helmholtz Association Initiative *Energy System 2050 - A Contribution of the Research Field Energy* and the grant No. VH-NG-1025; *STORM - Stochastics for Time-Space Risk Models* project of the Research Council of Norway (RCN) No. 274410, and the *E-ON Stipendienfonds*.
 
 ### References
 <sup>1</sup>Peng, C.-K., Buldyrev, S. V., Havlin, S., Simons, M., Stanley, H. E., & Goldberger, A. L. (1994). *Mosaic organization of DNA nucleotides*. [Physical Review E, 49(2), 1685–1689](https://doi.org/10.1103/PhysRevE.49.1685)\

contributions.md

@@ -1,5 +1,5 @@
 # Contributions
-`MDFA` has lived so far out of collaborations. It welcomes reviews and ideas from everyone. If you want to share your ideas or report a bug, open an [issue](https://github.com/LRydin/KramersMoyal/issues) here on GitHub, or contact me directly: [leonardo.rydin"at"gmail.com](mailto:leonardo.rydin@gmail.com)
+`MFDFA` has lived so far out of collaborations. It welcomes reviews and ideas from everyone. If you want to share your ideas or report a bug, open an [issue](https://github.com/LRydin/KramersMoyal/issues) here on GitHub, or contact me directly: [leonardo.rydin"at"gmail.com](mailto:leonardo.rydin@gmail.com)
 
 I'm are happy to have you help me in furthering this project, helping me improve the code and documentation, make it more user-friendly, broadening its applicability, or devising other methodologies of implementation.
 

docs/emd_detrending.rst

@@ -4,6 +4,14 @@ Employing Empirical Mode Decompositions for detrending
 `Empirical Mode Decomposition <https://en.wikipedia.org/wiki/Hilbert%E2%80%93Huang_transform>`_ (EMD), or maybe more correctly described, the Hilbert─Huang transform is a transformation analogous to a Fourier or Hilbert transform that decomposes a one-dimensional timeseries or signal into its Intrinsic Mode Functions (IMFs).
 For our purposes, we simply want to employ EMD to detrend a timeseries.
 
+.. warning::
+
+   To use this feature, you need to first install `PyEMD <https://github.com/laszukdawid/PyEMD>`_ (EMD-signal) with
+
+   ::
+
+      pip install EMD-signal
+
 Understanding :code:`MFDFA`'s :code:`EMD` detrender
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 

docs/installation.rst

@@ -1,14 +1,21 @@
 Installation
 ============
 
-For the moment the library is available from TestPyPI, so you can use
+:code:`MFDFA` is available from PyPI, so you can use
 
 ::
 
-   pip install -i https://test.pypi.org/simple/ MFDFA
+   pip install MFDFA
 
 Then on your favourite editor just use
 
 .. code:: python
 
    from MFDFA import MFDFA
+
+.. warning::
+   To use the extension to include Empirical Mode Decomposition detrending you will also need
+
+   ::
+
+      pip install EMD-signal

docs/license.rst

@@ -3,7 +3,7 @@ License
 
 MIT License
 
-Copyright (c) 2019-2020 Rydin
+Copyright (c) 2019-2021 Rydin
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

requirements.txt

@@ -1,2 +1 @@
 numpy
-EMD-signal

setup.py

@@ -5,18 +5,20 @@
 
 setuptools.setup(
     name="MFDFA",
-    version="0.3",
+    version="0.4",
     author="Leonardo Rydin Gorjao",
     author_email="leonardo.rydin@gmail.com",
     description="Multifractal Detrended Fluctuation Analysis in Python",
     long_description=long_description,
     long_description_content_type="text/markdown",
     url="https://github.com/LRydin/MFDFA",
     packages=setuptools.find_packages(),
+    install_requires = ["numpy"],
+    extras_require = {"EMD-signal": ["EMD-signal"]},
     classifiers=[
         "Programming Language :: Python :: 3",
         "License :: OSI Approved :: MIT License",
         "Operating System :: OS Independent",
     ],
-    python_requires='>=3.6',
+    python_requires='>=3.3',
 )

test/test_MFDFA.py

@@ -44,21 +44,6 @@ def test_MFDFA():
             assert edfa.ndim == 2, "Output is not 2 dimensional"
             assert dfa.shape[1] <= q.shape[0], "Output shape mismatch"
 
-            # Testing EMD extension
-            lag, dfa = MFDFA(X, lag = lag, q = q, order = 0,
-              stat = False, extensions = {'EMD': [0]})
-
-            assert dfa.ndim == 2, "Output is not 2 dimensional"
-            assert dfa.shape[1] <= q.shape[0], "Output shape mismatch"
-
-            # Testing EMD and eDFA extensions together
-            lag, dfa, edfa = MFDFA(X, lag = lag, q = q, order = 1,
-              stat = False, extensions = {'eDFA':True, 'EMD': [0]})
-
-            assert dfa.ndim == 2, "Output is not 2 dimensional"
-            assert edfa.ndim == 2, "Output is not 2 dimensional"
-            assert dfa.shape[1] <= q.shape[0], "Output shape mismatch"
-
             # Testing moving window
             lag, dfa = MFDFA(X, lag = lag, q = q, order = 1,
               stat = False, extensions = {'window': 5})

test/test_MFDFA_extras.py

@@ -0,0 +1,32 @@
+import numpy as np
+
+import sys
+sys.path.append("../")
+from MFDFA import MFDFA
+
+def test_MFDFA():
+    for N in [1000, 10000]:
+        for q_list in [1, 2, 6, 21]:
+
+            X = np.random.normal(size = N, loc = 0)
+            q = np.linspace(-10, 10, q_list)
+
+            lag = np.unique(
+                  np.logspace(
+                  0, np.log10(X.size // 4), 25
+                  ).astype(int) + 1
+                )
+
+            lag, dfa = MFDFA(X, lag = lag, q = q, order = 0,
+              modified = True, stat = False, extensions = {'EMD': [0]})
+
+            assert dfa.ndim == 2, "Output is not 2 dimensional"
+            assert dfa.shape[1] <= q.shape[0], "Output shape mismatch"
+
+            lag, dfa, edfa = MFDFA(X, lag = lag, q = q, order = 1,
+              modified = True, stat = False,
+              extensions = {'eDFA':True, 'EMD': [0]})
+
+            assert dfa.ndim == 2, "Output is not 2 dimensional"
+            assert edfa.ndim == 2, "Output is not 2 dimensional"
+            assert dfa.shape[1] <= q.shape[0], "Output shape mismatch"