Merge pull request #132 from cdt15/revert-128-master

Revert "Accelerate DirectLiNGAM by parallelising causal ordering on GPUs with CUDA"

lingam/direct_lingam.py

@@ -8,7 +8,6 @@
 from sklearn.utils import check_array
 
 from .base import _BaseLiNGAM
-from lingam_cuda import causal_order as causal_order_gpu
 
 
 class DirectLiNGAM(_BaseLiNGAM):
@@ -46,7 +45,7 @@ def __init__(
             * ``-1`` : No prior knowledge is available to know if either of the two cases above (0 or 1) is true.
         apply_prior_knowledge_softly : boolean, optional (default=False)
             If True, apply prior knowledge softly.
-        measure : {'pwling', 'kernel', 'pwling_fast'}, optional (default='pwling')
+        measure : {'pwling', 'kernel'}, optional (default='pwling')
             Measure to evaluate independence: 'pwling' [2]_ or 'kernel' [1]_.
         """
         super().__init__(random_state)
@@ -97,8 +96,6 @@ def fit(self, X):
         for _ in range(n_features):
             if self._measure == "kernel":
                 m = self._search_causal_order_kernel(X_, U)
-            elif self._measure == "pwling_fast":
-                m = self._search_causal_order_gpu(X_.astype(np.float64), U.astype(np.int32))
             else:
                 m = self._search_causal_order(X_, U)
             for i in U:
@@ -236,29 +233,6 @@ def _search_causal_order(self, X, U):
             M_list.append(-1.0 * M)
         return Uc[np.argmax(M_list)]
 
-    def _search_causal_order_gpu(self, X, U):
-        """Accelerated Causal ordering.
-
-        Parameters
-        ----------
-        X : array-like, shape (n_samples, n_features)
-            Training data, where ``n_samples`` is the number of samples
-            and ``n_features`` is the number of features.
-        U: indices of cols in X
-
-        Returns
-        -------
-        self : object
-            Returns the instance itself.
-        mlist: causal ordering
-        """
-        cols = len(U)
-        rows = len(X)
-
-        arr = X[:, np.array(U)]
-        mlist = causal_order_gpu(arr, rows, cols)
-        return U[np.argmax(mlist)]
-
     def _mutual_information(self, x1, x2, param):
         """Calculate the mutual informations."""
         kappa, sigma = param

lingam/utils/__init__.py

@@ -10,8 +10,6 @@
 import numpy as np
 import pandas as pd
 import semopy
-import subprocess
-
 from matplotlib import colors as mcolors
 from matplotlib.colors import is_color_like
 from sklearn import linear_model
@@ -1130,12 +1128,3 @@ def calculate_total_effect(adjacency_matrix, from_index, to_index, is_continuous
     total_effect = sum(effects)
 
     return total_effect
-
-def get_cuda_version():
-    try:
-        nvcc_version = subprocess.check_output(["nvcc", "--version"]).decode('utf-8')
-        print("CUDA Version found:\n", nvcc_version)
-        return True
-    except Exception as e:
-        print("CUDA not found or nvcc not in PATH:", e)
-        return False

requirements.txt

@@ -9,5 +9,3 @@ pygam
 matplotlib
 psy
 semopy
-#optional (install if gpu and cuda is available)
-# culingam==0.0.7