rllm.utils

Utils

download_url	Download the content of a URL to a specific folder.
download_google_url	Download the content of a Google Drive file to a specific folder.
extract_zip	Extract a zip archive to a specific folder.
sparse_mx_to_torch_sparse_tensor	Convert a scipy sparse matrix to a torch.sparse.Tensor.
is_torch_sparse_tensor	Return True if the input is a torch.sparse.Tensor.
get_indices	Get indices of non-zero elements from an adjacency matrix.
is_undirected	Check if the given adjacency matrix represents an undirected graph.
to_undirected	Convert the given adjacency matrix to an undirected graph representation by adding reverse edges.
adj_to_edge_index	Convert a sparse adjacency matrix to an edge index tensor.
seg_sum	Compute the sum of elements in data for each segment specified by segment_ids.
seg_softmax	Compute the segment-wise softmax scores of elements in data.
sort_edge_index	Sort the edge index.
index_to_ptr	Convert a sorted index tensor to a CSR/CSC pointer tensor.
lexsort	Perform an indirect stable sort using a sequence of keys.
remap_keys	Remap the keys of the input dictionary using a mapping.
CastMixin	Support cast init from tuple or dict input.