tutorials.examples.train_discreteebm¶
The goal of this script is to reproduce some of the published results on the HyperGrid environment. Run one of the following commands to reproduce some of the results in [Trajectory balance: Improved credit assignment in GFlowNets](https://arxiv.org/abs/2201.13259)
python train_hypergrid.py –ndim 4 –height 8 –R0 {0.1, 0.01, 0.001} –tied {–uniform} –loss {TB, DB} python train_hypergrid.py –ndim 2 –height 64 –R0 {0.1, 0.01, 0.001} –tied {–uniform} –loss {TB, DB}
And run one of the following to reproduce some of the results in [Learning GFlowNets from partial episodes for improved convergence and stability](https://arxiv.org/abs/2209.12782) python train_hypergrid.py –ndim {2, 4} –height 12 –R0 {1e-3, 1e-4} –tied –loss {TB, DB, SubTB}
Attributes¶
Functions¶
|
Module Contents¶
- tutorials.examples.train_discreteebm.DEFAULT_SEED = 4444¶
- tutorials.examples.train_discreteebm.main(args)¶
- tutorials.examples.train_discreteebm.parser¶