o i@sjddlZddlZzddlZdZWn eydZYnw  dddZeddd d Zed d d dZ dS)NTFweakc Cs~tstd|}|dvrtd|sd}|jdkrtdtjj|s-tjj |}|j \}}||kr:td|j dkrJ|t j ||jjd fS|d krft j||jjd }tj|j|jd ||j |d n#||j7}tjj|sytjj |}tjd d |j|jd d d d\}}t jd|jjd }t j|f|jjd } t||| |dt|d} |s| St| t | d | || |fS)aVAnalyzes the connected components of a sparse graph Args: csgraph (cupy.ndarray of cupyx.scipy.sparse.csr_matrix): The adjacency matrix representing connectivity among nodes. directed (bool): If ``True``, it operates on a directed graph. If ``False``, it operates on an undirected graph. connection (str): ``'weak'`` or ``'strong'``. For directed graphs, the type of connection to use. Nodes i and j are "strongly" connected only when a path exists both from i to j and from j to i. If ``directed`` is ``False``, this argument is ignored. return_labels (bool): If ``True``, it returns the labels for each of the connected components. Returns: tuple of int and cupy.ndarray, or int: If ``return_labels`` == ``True``, returns a tuple ``(n, labels)``, where ``n`` is the number of connected components and ``labels`` is labels of each connected components. Otherwise, returns ``n``. .. seealso:: :func:`scipy.sparse.csgraph.connected_components` zpylibcugraph is not available)rstrongz%connection must be 'weak' or 'strong'rz graph should have two dimensionszgraph should be a square arrayr)dtyperN)offsetsindicesweights num_verts num_edgeslabelsF)resource_handlegraphrrrr do_expensive_check))size)pylibcugraph_available RuntimeErrorlower ValueErrorndimcupyxscipysparseisspmatrix_csr csr_matrixshapennzcupyarangerrempty pylibcugraphstrongly_connected_componentsindptrTweakly_connected_componentszeros_cupy_count_componentsint_cupy_adjust_labelssort) csgraphdirected connection return_labelsmm1r _count root_labelsnr4b/home/ubuntu/veenaModal/venv/lib/python3.10/site-packages/cupyx/scipy/sparse/csgraph/_traversal.pyconnected_components sT        r6z4raw I labels, raw int32 count, raw int32 root_labelsz int j = i; while (j != labels[j]) { j = labels[j]; } if (j != i) { labels[i] = j; } else { int k = atomicAdd(&count[0], 1); root_labels[k] = i; } r&z&int32 n_root_labels, raw I root_labelszI labelsav int cur_label = labels; int j_min = 0; int j_max = n_root_labels - 1; int j = (j_min + j_max) / 2; while (j_min < j_max) { if (cur_label == root_labels[j]) break; if (cur_label < root_labels[j]) { j_max = j - 1; } else { j_min = j + 1; } j = (j_min + j_max) / 2; } labels = j; r()TrT) rcupyx.scipy.sparserr rModuleNotFoundErrorr6ElementwiseKernelr&r(r4r4r4r5s.  J