-
With the [current code](https://github.com/Learning-and-Intelligent-Systems/predicators/blob/f3b9bf125fe8c1d861c678010626d3b6570c47d4/predicators/planning.py#L1163) in `_sesame_plan_with_fast_downward…
-
It is time now to discuss the grounder selector.
I faced some issues with some domains due to the grounder's automatic selection. For context, UP searches which grounder engines are installed and d…
-
Considering how much we call fast downward it would be good to speed up the process. It appears that he translation step will take up most of the time. Ideally this could be improved, solutions migh…
lotu updated
12 years ago
-
It is necessarily to evaluate the different search methods and heuristics we can use with fast downward to improve our search. Currently astar(blind()) is used but this should be improved upon. This…
lotu updated
12 years ago
-
@ctpelok77 there seems to be a significant scaling issue when negative preconditions are introduced under not all but certain circumstances (not sure which!!).
Here the only difference between the…
-
When running the PDDL test using the following planner configuration
```json
{
"kind": "COMMAND",
"path": "/Users/victor.paleologue/Code/Palaio/HTN/downward/fast-do…
-
When I play the file `json_2.1.1/valid_seen/pick_two_obj_and_place-Spatula-None-Drawer-13/trial_T20190908_183331_203888`, the environment update takes nearly 6 minutes. Is it normal? Can I estimate th…
-
**I am encountering an issue while using your STRIPS-HGN tool for training, and I would appreciate your assistance. The specific problem is as follows:
During the execution of the training script, …
-
### 🎮 Game Request
This is a platform game set in a tower, where the player's goal is to jump from one "floor" to the next and go as high as possible without falling. The higher the player's characte…
-
A specialized option of the ```Run``` class (e.g., ```FastDownwardRun```) could make life easier for the user when defining a FD run by providing an interface where the planner, the PDDL files, etc. c…