Learning-and-Intelligent-Systems/llm4pddl

Issues

• Could you please add an instruction to help us understand how to use the codebase?

  #172 opened 2 years ago by JiaxinQin0814
  3

• always randomize the names of objects

  #103 opened 2 years ago by tomsilver
  0

• autogressive prompting + name randomization doesn't update prompt correctly

  #169 opened 2 years ago by tomsilver
  0

• change analyze results node counting to average over only tasks solved by all approaches

  #130 opened 2 years ago by tomsilver
  0

• Speed up checks

  #137 opened 2 years ago by tomsilver
  0

• Potentially unneeded queries being made when importing tokenizer

  #134 opened 2 years ago by Alchemy5
  0

• make breaking vs continuing after an invalid action a flag

  #131 opened 2 years ago by tomsilver
  0

• make sure that dressed problems are different enough

  #86 opened 2 years ago by tomsilver
  0

• fix analyze_results.py in case where experiment_id is not defined, or different from standard format

  #118 opened 2 years ago by tomsilver
  0

• implement timeout for all of planning, and for LLM

  #122 opened 2 years ago by tomsilver
  0

• add eval task num offset

  #110 opened 2 years ago by tomsilver
  0

• switch parentheses format in prompts

  #116 opened 2 years ago by tomsilver
  0

• try blocks where the names of the blocks have integers

  #117 opened 2 years ago by tomsilver
  0

• try blocks with data from manually designed generalized policy

  #102 opened 2 years ago by tomsilver
  0

• implement PG3 baseline

  #112 opened 2 years ago by tomsilver
  0

• add manual training problems

  #101 opened 2 years ago by tomsilver
  0

• add training index offset

  #98 opened 2 years ago by tomsilver
  0

• add option to do multiple queries for multi-* approaches

  #94 opened 2 years ago by tomsilver
  0

• consider removing domains that have per-problem domain files

  #88 opened 2 years ago by tomsilver
  0

• consider switching eval problems to be hardest ones, rather than next-easiest ones

  #91 opened 2 years ago by tomsilver
  0

• data augmentation: only keep tasks whose plans use all the operators

  #84 opened 2 years ago by tomsilver
  0

• remove `get_pyperplan_benchmark_task` and `get_custom_benchmark_task`

  #81 opened 2 years ago by tomsilver
  0

• add cached problem_str and domain_str to Task and refactor accordingly

  #72 opened 2 years ago by tomsilver
  2

• create AB testing script

  #69 opened 2 years ago by tomsilver
  0

• add constant objects to prompt

  #56 opened 2 years ago by tomsilver
  0

• save and load datasets

  #62 opened 2 years ago by tomsilver
  0

• analyze the problem sizes for the pyperplan benchmark tasks

  #57 opened 2 years ago by tomsilver
  1

• change the llm cache file names to not use hashing

  #40 opened 2 years ago by tomsilver
  0

• logs don't get written when locally launching configs

  #31 opened 2 years ago by tomsilver
  0

• compare against FD with hff and gbfs

  #22 opened 2 years ago by tomsilver
  4