Pinchuk Maya
Pinchuk Maya
The current percent of coverage is ~74%, which is relatively small and allows to make more thoughtless and untested changes. Even though some functionality can be covered with integration tests,...
Make a research on existing benchmarks for graph structure learning and automate the use of it. For example, [this article](https://arxiv.org/abs/2310.05163v1) ([github](https://github.com/GSL-Benchmark/GSLB)) must be considered.
While launching experiments for `GOLEM paper`, it was noticed that every 3-4 launches `GOLEM `finishes optimization early because it cannot generate enough individuals for `gnp `and `grid2d `graph types. However,...
Experiments to compare random GOLEM with MAB should be conducted: - On the simple task - On the graph search task Convergence, mutations probability and population diversity must be analyzed
It is important to filtrate individuals frequently since now the same structurally identical individuals are rapidly accumulating. However, too frequent filtering can significantly slow down the evolution process. The task...
Experiments to compare random GOLEM with Contextual MAB should be conducted: On the simple task On the graph search task Convergence, mutations probability and population diversity must be analyzed. Mutation...
Now labels for such edges are placed in the middle of edge if it was straight:  Label should be placed in the middle of **curved** edge
pip install git+https://github.com/aimclub/FEDOT.git@59774e6a7fadb5b2140a0f1c513f8173483a418e pip uninstall thegolem -y pip install git+https://github.com/aimclub/GOLEM.git@46b53a49e5651698bfef82dccccfee65f2a1c11f **NOW** pip install git+https://github.com/aimclub/FEDOT.git@e6d2740d4e80b3c57fd10894431ebb4e516bf320 pip uninstall thegolem -y pip install git+https://github.com/aimclub/GOLEM.git@a6f2b903f4f57919ff137156588374b2e2285eca