This software primarily implements our two proposed signed whole graph embedding methods: SG2V (Signed Graph2vec) and WSGCN (Whole Signed Graph Convolutional Networks). The software can be used to learn the representations of whole signed graphs, and perform classification tasks based on the embeddings extracted with our method. In addition, it can reproduce our experiments described in the linked paper. In particular, it compares the performance of our method with three alternatives from the literature: SiNE, which handles signed graphs, but only to represent individual vertices, and not the whole graph; Graph2vec which handles whole-graphs, but only for unsigned graphs; and SGCN which handles signed graphs, but only to represent individual vertices, and not the whole graph.