graphkit-learn: A Python library for graph kernels based on linear patterns

• Graph kernels bridge the gap between machine learning and data encoded as graphs. based on linear patterns provide good performance. A Python library for efficient computation of graph is proposed. Three strategies are provided to reduce computational complexity. Experiments synthesized real-world datasets show relevance library. This paper presents graphkit-learn , first patterns, able address various types thoroughly implemented, each with specific computing methods, well two well-known non-linear comparative analysis. Since complexity an Achilles’ heel kernels, we several this critical issue, including parallelization, trie structure, FCSP method that extend other edge comparison. All proposed save orders magnitudes time memory usage. Moreover, all can be simply computed a single statement, thus appealing researchers practitioners. For convenience use, advanced model selection procedure both regression classification problems. 11 benchmark