bs-python-utils 0.4__tar.gz → 0.4.2__tar.gz

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: bs-python-utils
-Version: 0.4
+Version: 0.4.2
 Summary: my Python utilities
 Home-page: https://github.com/bsalanie/bs-python-utils
 Author: Bernard Salanie
@@ -13,6 +13,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Requires-Dist: altair (>=4.2.2,<5.0.0)
 Requires-Dist: altair-saver (>=0.5.0,<0.6.0)
+Requires-Dist: griffe (>=0.32.3,<0.33.0)
 Requires-Dist: matplotlib (>=3.7.1,<4.0.0)
 Requires-Dist: scikit-learn (>=1.2.2,<2.0.0)
 Requires-Dist: seaborn (>=0.12.2,<0.13.0)
@@ -36,7 +37,10 @@ My Python utilities.
 -   **Documentation** <https://bsalanie.github.io/bs-python-utils/>
 
 ### Release notes
-
+#### 0.4.2 (August 2, 2023)
+Updated the docs.
+#### 0.4.1 (July 23, 2023)
+Only print if verbose in bivariate quantiles.
 #### 0.4 (July 22, 2023)
 Added bivariate quantiles and ranks à la optimal transportation.
 #### 0.3 (July 21, 2023)

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/README.md

@@ -12,7 +12,10 @@ My Python utilities.
 -   **Documentation** <https://bsalanie.github.io/bs-python-utils/>
 
 ### Release notes
-
+#### 0.4.2 (August 2, 2023)
+Updated the docs.
+#### 0.4.1 (July 23, 2023)
+Only print if verbose in bivariate quantiles.
 #### 0.4 (July 22, 2023)
 Added bivariate quantiles and ranks à la optimal transportation.
 #### 0.3 (July 21, 2023)

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/bs_python_utils/bivariate_quantiles.py

@@ -122,6 +122,7 @@ def bivariate_ranks_v(
     Args:
         y: the observations, an `(n,2)` matrix
         v: the converged values of the weights, an `n`-vector
+        n_nodes: the number of nodes for Chebyshev integration
         presorted: if `True`, then `y` and `v` are sorted by increasing `y[:, 1]`.
 
     Returns:
@@ -234,12 +235,14 @@ def solve_for_v_(y: np.ndarray, n_nodes: int = 32, verbose: bool = False) -> np.
     argsog = [y_sorted, u1_nodes, u1_weights, verbose]
 
     res = minimize_free(obj_, grad_, v0, args=argsog)
-    print_optimization_results(res, "Minimizing over v")
+    if verbose:
+        print_optimization_results(res, "Minimizing over v")
 
     if not res.success:
         bs_error_abort("Problem! the optimization failed.")
     vstar = res.x
-    print(f"The final gradient is close to 0: error {npmaxabs(res.jac)}")
+    if verbose:
+        print(f"The final gradient over v is close to 0: error {npmaxabs(res.jac)}")
     vstar1_sorted = np.append(vstar, -np.sum(vstar))
 
     # revert to original order
@@ -251,12 +254,12 @@ def solve_for_v_(y: np.ndarray, n_nodes: int = 32, verbose: bool = False) -> np.
 
 def bivariate_quantiles(
     y: np.ndarray, u: np.ndarray, n_nodes: int = 32, verbose: bool = False
-):
+) -> np.ndarray:
     """computes the bivariate quantiles of `y` at the quantiles `u`
 
     Args:
         y: the observations, an `(n, 2)` matrix
-        u_qtiles: the quantiles at which to compute the bivariate quantiles,
+        u: the quantiles at which to compute the bivariate quantiles,
             an `(m, 2)` matrix
         n_nodes: the number of nodes to use for the quadrature
         verbose: if `True`, print some information
@@ -268,7 +271,9 @@ def bivariate_quantiles(
     return bivariate_quantiles_v(y, u, v)
 
 
-def bivariate_ranks(y: np.ndarray, n_nodes: int = 32, verbose: bool = False):
+def bivariate_ranks(
+    y: np.ndarray, n_nodes: int = 32, verbose: bool = False
+) -> np.ndarray:
     """computes the bivariate ranks of `y`
 
     Args:

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/bs_python_utils/bs_sparse_gaussian.py

@@ -1,5 +1,11 @@
 """
-sets up sparse integration over a Gaussian
+Sets up sparse integration over a Gaussian, given text files that contain rescaled Gauss-Hermite nodes and weights.
+
+These files must be named `GHsparseGrid{ndims}prec{iprec}.txt`, where `ndims` is the number of dimensions of integration 
+and `iprec` is a precision level that must be 9, 13, or (most precise) 17. The file must have `(ndims+1) columns, 
+with the weights in the first column.
+
+The nodes and weights are rescaled so that `f(nodes) @ weights` approximates `Ef(X)` for `X` an `N(0,I)` variable.
 """
 from pathlib import Path
 
@@ -15,11 +21,12 @@ def setup_sparse_gaussian(
     """
     get nodes and weights for sparse integration Ef(X) with X = N(0,1) in `ndims` dimensions
 
-    usage: nodes, weights = setup_sparse_gaussian(mdims, iprec); intf = f(nodes) @ weights
+    usage: nodes, weights = setup_sparse_gaussian(mdims, iprec); integral_f = f(nodes) @ weights
 
     Args:
         ndims: number of dimensions (1 to 5)
         iprec: precision (must be 9, 13, or 17)
+        GHsparsedir: the name of a directory that contains nodes and weights
 
     Returns:
         a pair of  arrays `nodes` and `weights`;
@@ -36,9 +43,17 @@ def setup_sparse_gaussian(
         )
 
     if ndims in [1, 2, 3, 4, 5]:
+        if not GHdir.exists():
+            bs_error_abort("I did not find the directory with the nodes/weights files.")
         grid = np.loadtxt(GHdir / f"GHsparseGrid{ndims}prec{iprec}.txt")
-        weights = grid[:, 0]
-        nodes = grid[:, 1:]
+
+        print(f"{grid.shape=}")
+        if ndims == 1:
+            weights = grid[:, 0]
+            nodes = grid[:, 1]
+        else:
+            weights = grid[:, 0]
+            nodes = grid[:, 1:]
         return nodes, weights
     else:
         bs_error_abort(

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/bs_python_utils/bsstats.py

@@ -16,7 +16,6 @@ from bs_python_utils.bsnputils import (
     check_matrix,
     check_vector,
     check_vector_or_matrix,
-    make_lexico_grid,
 )
 from bs_python_utils.bssputils import spline_reg
 from bs_python_utils.bsutils import bs_error_abort
@@ -416,37 +415,3 @@ def estimate_pdf(
         if weights is not None:
             f_x *= weights / np.mean(weights)
     return cast(np.ndarray, f_x)
-
-
-def estimate_densities_at_quantiles(
-    X: np.ndarray, qtiles: np.ndarray
-) -> tuple[np.ndarray, np.ndarray] | tuple[np.ndarray, np.ndarray, np.ndarray]:
-    """estimate densities of margins at prespecified quantiles (Silverman rule)
-    and the joint density at each vector of these quantiles
-
-    Args:
-        X: `n`-vector or `(n, nx)`-matrix
-        qtiles: vector of `nq` numbers between 0 and 1
-
-    Returns: 
-        if `X` is a matrix, the `({nq}^{nx}, {nx})` matrices of estimated margin densities \
-     and the `{nq}^{nx}` vector of the joint density on the lexicographic grid of quantiles;
-        if `X` is a vector, the `nq`-vector of the density at the quantiles, twice
-    """
-    ndims_X = check_vector_or_matrix(X, "estimate_densities_")
-    if ndims_X == 1:
-        f_X = estimate_pdf(X, np.quantile(X, qtiles))
-        return f_X, f_X
-    else:
-        nx = X.shape[1]
-        nq = qtiles.size
-        f_X_k = np.zeros((nq, nx))
-        nodes_mat = np.zeros((nq, nx))
-        for i_x in range(nx):
-            X_ix = X[:, i_x]
-            nodes_mat[:, i_x] = np.quantile(X_ix, qtiles)
-            f_X_k[:, i_x] = estimate_pdf(X_ix, nodes_mat[:, i_x])
-        f_margins = make_lexico_grid(f_X_k)
-        values_X = make_lexico_grid(nodes_mat)
-        f_X = estimate_pdf(X, values_X)  # joint density
-        return f_margins, f_X, values_X

{bs_python_utils-0.4 → bs_python_utils-0.4.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "bs_python_utils"
-version = "0.4"
+version = "0.4.2"
 description = "my Python utilities"
 authors = ["Bernard Salanie <bsalanie@columbia.edu>"]
 repository = "https://github.com/bsalanie/bs-python-utils"
@@ -19,6 +19,7 @@ seaborn = "^0.12.2"
 scikit-learn = "^1.2.2"
 statsmodels = "^0.13.5"
 vega-datasets = "^0.9.0"
+griffe = "^0.32.3"
 
 [tool.poetry.group.dev.dependencies]
 pytest = "^7.2.0"