A few experimental implementations #67
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* update README with notes for installing box2d-py * update README
* feat: add render toggle via Hydra * feat: Add one of prbench's envs * style(black): format run_experiment.py * fix(lint): move the lazy import, resolve warnings * feat: add registery class for supporting multiple environments (prbench, ggg) * refactor: move gym_to_gymnasium and wrappers to utils directory * feat: add the visualization logic for ggg * fix: minor updates * feat: add test files for envs * feat: add tests for GymToGymnasium and wrappers; fix adapter compatibility with step function * fix: removed the extra prints * fix: modifying a local import to an absoulte one * fix: add type annotation for ggg and pr_bench function * fix: renamed ggg and pr_bench file and function names * fix: use built-in types and | None for type annotations * refatcor: change gym_to_gymnasium's step function * feat: add test for gym_to_gymnasium and wrappers * fix: use OmegaConf config in env tests * fix: update gitignore * fix: type annotation issues
* Change BaseApproach based on the needs for LPP * Add policies directory, including lpp_policy.py * Add grammar and dsl logic * Add demonstration logic * Remove all the added files for LPP * Add minimal LPP implementation with primitives and StateActionProgram * Fix type annotations * Remove classes from tests * Fix the absolute import * Make the StateActionProgram class more generic * Add tests for eval_context addition to StateActionProgram * Change Dict to dict
* Change BaseApproach based on the needs for LPP * Add policies directory, including lpp_policy.py * Add grammar and dsl logic * Add demonstration logic * Remove all the added files for LPP * Add minimal LPP implementation with primitives and StateActionProgram * Fix type annotations * Remove classes from tests * Fix the absolute import * Make the StateActionProgram class more generic * Add tests for eval_context addition to StateActionProgram * Change Dict to dict * Add all primitives * Apply a few fixes
* Change BaseApproach based on the needs for LPP * Add policies directory, including lpp_policy.py * Add grammar and dsl logic * Add demonstration logic * Remove all the added files for LPP * Add minimal LPP implementation with primitives and StateActionProgram * Fix type annotations * Remove classes from tests * Fix the absolute import * Make the StateActionProgram class more generic * Add tests for eval_context addition to StateActionProgram * Change Dict to dict * Add all primitives * Apply a few fixes * Add grammar builder --------- Co-authored-by: Tom Silver <[email protected]>
* Update README * Refactor code to include different primitives for different envs * Finalize provider-specific dsl creation + tests
* Update README * Refactor code to include different primitives for different envs * Finalize provider-specific dsl creation + tests * Remove the LPP Implementation and tests
* Update README * Refactor code to include different primitives for different envs * Finalize provider-specific dsl creation + tests * Remove the LPP Implementation and tests * Add grid_v1(primitives for ggg), and the DSL base classs * Fix minor stuff * Add test_core.py * Add primitives_test * Add test for grid_v1 eval * Change a name
* Update README * Refactor code to include different primitives for different envs * Finalize provider-specific dsl creation + tests * Remove the LPP Implementation and tests * Add grid_v1(primitives for ggg), and the DSL base classs * Fix minor stuff * Add test_core.py * Add primitives_test * Add test for grid_v1 eval * Change a name * Add DSLLearner base and oracle classes * Add test for OracleLearner * Made DSLLearner more type-specific * Change in pyproject.toml
* program generators * tests pass but messy * wip * Fix type-annotations and minor issues * Add a test with input variables (x) * Add ggg grammar + env class that contains getType * Add test + final adjustments --------- Co-authored-by: Tom Silver <[email protected]>
* Add data folder with demo_types and collect method + tests * Add tiny test for demo_types * Keep one dataclass in demo_types.py * Remove if block and replace it with an assertion
* Add data folder with demo_types and collect method + tests * Add tiny test for demo_types * Keep one dataclass in demo_types.py * Remove if block and replace it with an assertion * Add grid_experts.py including expert policies for grid games * Add test for grid_experts * Change minor * Fix formatting
* Change BaseApproach class * Add expert_approch class + move the expert_grids logic * Add test for expert approach * Fix ci checks * Fix type annotations * Add grid_experts test * Add a test to connect ExpertApproach class with the grid_experts and demo_collection * Fix a comment * Fix a comment
* Add dataset file * Change mypy sensitivity and fix many files annotations * Add tests for dataset.py * Add pipline test for dataset-creation * Fix demo failure * Modify Trajectory dataclass type and apply changes to dataset.py
* Update README * Update dependency in readme * Add isntructions on implementing a custom env * Fix format * Fix format
* Add dataset file * Change mypy sensitivity and fix many files annotations * Add tests for dataset.py * Add pipline test for dataset-creation * Fix demo failure * Modify Trajectory dataclass type and apply changes to dataset.py * Return log_priors * Fix types * Add StateActionProgram class * Add test for StateActionProgram * Pass dsl_functions to StateActionProgram * Apply a minor change
* add gym maze to environment registry * implement outerworld to maze environment * add maze npy file for maze_env * add custom maze to registry * clean up files * address linting errors * remove pylint error * remove lint errors * add maze env test file * add maze env test file * fix autoformat error * address PR comments
* Add decision_tree_learner functions * Add tests * Fix str and rep functions' return types (StateActionProgram class)
* Add lpp_policy initial draft * Update lpp policy class * Add tests for lpp policy class * Make LPPPolicy class more type generic
* Add decision_tree_learner functions * Add tests * Fix str and rep functions' return types (StateActionProgram class) * Add plp_likelihood files * Add plp_likelihood.py file * Add tests for PLP likelihood computation * Fix type annotations * Add particle_utils.py + test
… merge_results step
… merge_results step
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I can go over the changes in our meeting tomorrow! A bunch of files changed, but nothing big was added, it's mostly experiment-setup stuff and a few spots where I updated arg types and fixed the tests to match.