gfdl 0.1.0__py3-none-any.whl

gfdl/weights.py

@@ -0,0 +1,378 @@
+"""
+Weight functions for Gradient Free Deep Learning estimators.
+"""
+
+import numpy as np
+
+
+def zeros(d, h, **kwargs):
+    """
+    The weight function setting all weights to zero.
+
+    This function is useful to test out the effect of
+    the data features in isolation.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    Returns
+    -------
+    ndarray or scalar
+        All zeros.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+    """
+    return np.zeros((h, d))
+
+
+def uniform(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from uniform distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the uniform distribution between ``[0, 1)``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Other keyword arguments. Placeholder for exposing distribution
+        parameters later on.
+    """
+    return rng.uniform(0, 1, (h, d))
+
+
+def range(d, h, **kwargs):
+    """
+    The weight function setting weights to a normalized np.arange.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    Returns
+    -------
+    ndarray or scalar
+        Set the weights to normalized np.arange over the range ``[0, d*h)``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+    """
+    s = np.arange(d * h)
+    s = np.subtract(s, np.mean(s))
+    s /= np.std(s)
+    s = np.nan_to_num(s)
+    return s.reshape(h, d)
+
+
+def he_uniform(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from He uniform distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the He uniform distribution between
+        ``[sqrt(6/h), sqrt(6/h))``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://faroit.com/keras-docs/2.0.0/initializers/#he_uniform
+    """
+
+    limit = np.sqrt(6 / h)
+    return rng.uniform(-limit, limit, (h, d))
+
+
+def lecun_uniform(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from Lecun uniform distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the Lecun uniform distribution between
+        ``[sqrt(3/h), sqrt(3/h))``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://faroit.com/keras-docs/2.0.0/initializers/#lecun_uniform
+    """
+
+    limit = np.sqrt(3 / h)
+    return rng.uniform(-limit, limit, (h, d))
+
+
+def glorot_uniform(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from Glorot uniform distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the Glorot uniform distribution between
+        ``[-sqrt(6/(d+h)), sqrt(6/(d+h)))``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://faroit.com/keras-docs/2.0.0/initializers/#glorot_uniform
+    """
+
+    fan_avg = 0.5 * (d + h)
+    limit = np.sqrt(3 / fan_avg)
+    return rng.uniform(-limit, limit, (h, d))
+
+
+def normal(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from normal distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the normal distribution with
+        mean ``0`` and standard deviation ``1``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Other keyword arguments. Placeholder for exposing distribution
+        parameters later on.
+    """
+    return rng.normal(0, 1, (h, d))
+
+
+def he_normal(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from He normal distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the He normal distribution with
+        mean ``0`` and standard deviation ``sqrt(2/h)``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://faroit.com/keras-docs/2.0.0/initializers/#he_normal
+    """
+
+    var = np.sqrt(2 / h)
+    return rng.normal(0, var, (h, d))
+
+
+def lecun_normal(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from Lecun normal distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the Lecun normal distribution
+        with mean ``0`` and standard deviation ``sqrt(1/h)``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://www.tensorflow.org/api_docs/python/tf/keras/initializers/LecunNormal
+    """
+
+    var = 1 / np.sqrt(h)
+    return rng.normal(0, var, (h, d))
+
+
+def glorot_normal(d, h, *, rng, **kwargs):
+    """
+    The weight function returning samples drawn from Glorot normal distribution.
+
+    Parameters
+    ----------
+    d : int
+        Number of features in a sample or number of neurons in the previous
+        hidden layer.
+
+    h : int
+        Number of neurons in the current hidden layer.
+
+    rng : np.random.Generator
+        A NumPy random number generator instance.
+
+    Returns
+    -------
+    ndarray or scalar
+        Draw samples from the Glorot normal distribution with
+        mean ``0`` and standard deviation ``sqrt(2/(d+h))``.
+
+    Other Parameters
+    ----------------
+    **kwargs : dict
+        Needed for keyword arguments and compatibility with other weight function apis
+        but not relevant for this function.
+
+    Notes
+    -----
+    https://faroit.com/keras-docs/2.0.0/initializers/#glorot_normal
+    """
+
+    fan_avg = 0.5 * (d + h)
+    var = np.sqrt(1 / fan_avg)
+    return rng.normal(0, var, (h, d))
+
+
+WEIGHTS = {
+    "zeros": zeros,
+    "uniform": uniform,
+    "range": range,
+    "normal": normal,
+    "he_uniform": he_uniform,
+    "lecun_uniform": lecun_uniform,
+    "glorot_uniform": glorot_uniform,
+    "he_normal": he_normal,
+    "lecun_normal": lecun_normal,
+    "glorot_normal": glorot_normal,
+}
+
+
+def resolve_weight(weight):
+    # numpydoc ignore=GL08
+    name = weight.strip().lower()
+    try:
+        w = WEIGHTS[name]
+    except KeyError as e:
+        allowed = sorted(WEIGHTS.keys())
+        raise ValueError(
+            f"weight scheme='{weight}' is not supported; choose from {allowed}"
+        ) from e
+    return w

gfdl-0.1.0.dist-info/METADATA

@@ -0,0 +1,33 @@
+Metadata-Version: 2.4
+Name: gfdl
+Version: 0.1.0
+Summary: Gradient Free Deep Learning (GFDL) networks including single and multi layer random vector functional link (RVFL) networks and extreme learning machines (ELMs)
+Project-URL: source, https://github.com/lanl/GFDL
+Project-URL: download, https://github.com/lanl/GFDL/releases
+Project-URL: tracker, https://github.com/lanl/GFDL/issues
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Science/Research
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: Scientific/Engineering
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Operating System :: POSIX
+Classifier: Operating System :: Unix
+Classifier: Operating System :: MacOS
+Requires-Python: >=3.12
+License-File: COPYING
+Requires-Dist: numpy<2.7,>=2.0.0
+Requires-Dist: scikit-learn<1.11,>=1.5.0
+Requires-Dist: scipy<1.20,>=1.13.0
+Requires-Dist: packaging<27.0,>=24.0
+Provides-Extra: test
+Requires-Dist: ucimlrepo; extra == "test"
+Requires-Dist: pytest; extra == "test"
+Requires-Dist: pytest-cov; extra == "test"
+Dynamic: license-file

gfdl-0.1.0.dist-info/RECORD

@@ -0,0 +1,12 @@
+gfdl/__init__.py,sha256=RZGxuS1S6CtWVb78TfscnwBXPNGTdT55DHVFiO5plFc,1156
+gfdl/activations.py,sha256=V5_49AEV8rhGHjk56Lml-bUqiN0_kKqWs3Qsl_UJscw,3483
+gfdl/model.py,sha256=d-rhxJi454yqW0Cz4bxBrjxX-F6XTfkwRo_mYReaTzg,28758
+gfdl/weights.py,sha256=PFcO2C-IppP3b_EvukwPIWwqyJROpWje-JsGz9HIbe8,9397
+gfdl/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+gfdl/tests/test_model.py,sha256=FtUAVHlDAXB6kidhiVTPqSvkclRwlhpkHdEpwhK_glU,20304
+gfdl/tests/test_regression.py,sha256=00FSkqw1WFVhoi3APtypUaNYmKZjgqvpwl3rpHsP6J4,6465
+gfdl-0.1.0.dist-info/licenses/COPYING,sha256=SyvHPYvDWhkWHrNALv0ppjB9yJhLi9mMRdemPF9wYPA,1535
+gfdl-0.1.0.dist-info/METADATA,sha256=SiATGBdeGXFbFmtb5uDUqA08LCPvo3sypnLn5FPnjLw,1438
+gfdl-0.1.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+gfdl-0.1.0.dist-info/top_level.txt,sha256=Sdh-XBQCopJ8FC1mqqCi5dxuzGibAGEPUCTtzltZ84c,5
+gfdl-0.1.0.dist-info/RECORD,,

gfdl-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.10.2)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

gfdl-0.1.0.dist-info/licenses/COPYING

@@ -0,0 +1,29 @@
+BSD 3-Clause License
+
+Copyright (c) 2025-2026 Los Alamos National Laboratory.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+* Neither the name of the copyright holder nor the names of its
+  contributors may be used to endorse or promote products derived from
+  this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

gfdl-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+gfdl