Tivadar Danka

Hierarchical sampling for active learning

1

Implement the hierarchical sampling algorithm from [Dasgupta and Hsu paper](https://dl.acm.org/citation.cfm?id=1390183).

Query synthesis

3

Active learning not only works in pool-based or stream-based setting, it can *generate* examples which can be queried for labels. This is called query synthesis. (See [this paper](https://www.sciencedirect.com/science/article/pii/S0925231214008145) for further...

Support for `torch.Tensor` data types

2

The learners should support `torch.Tensor` datatypes, see: #109