Skip to content

Conversation

@mpragnay
Copy link

  • Cleaned up the processing branch to produce a set of Carla maps with valid objects, intended for upload to HuggingFace.

  • Added all 8 Carla towns generated via py123d and custom pyxodr processing script.

  • Validated Z-coordinates to ensure agents initialize correctly on complex geometries, including overpasses and underpasses.

  • Added setup instructions for the pyxodr submodule dependency, users can generate custom agents with specific parameters (e.g., distant goals, non-zero initial velocity).

@greptile-apps
Copy link

greptile-apps bot commented Dec 17, 2025

Greptile Overview

Greptile Summary

Cleaned up Carla map processing pipeline and added 8 validated town maps with procedurally generated agent trajectories.

Key Changes:

  • Replaced unstable process_carla_roads.py with generate_carla_agents.py featuring argument parsing, AABB collision detection, and configurable trajectory generation
  • Added pyxodr submodule dependency for robust XODR processing
  • Generated and validated all 8 Carla towns (Town01-07, Town10HD) via py123d processing with correct Z-coordinate handling for complex geometries
  • Added visualization tools (plot.py) and test scripts (JunctionOutlineTest.py) for debugging junction polygonization
  • Updated documentation with setup instructions and usage examples

Issues Found:

  • Typo in README.md line 197: "genrated" should be "generated"

Confidence Score: 4/5

  • This PR is safe to merge with one minor typo fix needed
  • Score reflects well-structured code refactoring with proper collision detection, Z-coordinate validation for complex geometries, and comprehensive documentation. The main concern is the typo in documentation and ensuring the pyxodr submodule is properly initialized after checkout. The code demonstrates good practices with AABB collision detection, configurable parameters via argparse, and proper trajectory generation logic.
  • README.md needs typo fix on line 197. Verify that external/pyxodr submodule initializes correctly after clone.

Important Files Changed

File Analysis

Filename Score Overview
.gitmodules 5/5 Added pyxodr submodule from mpragnay/pyxodr GitHub repository for XODR processing
requirements-dev.txt 5/5 Added development dependency for pyxodr submodule in editable mode
README.md 4/5 Added documentation for generating Carla agent trajectories with setup instructions and usage examples
data_utils/carla/generate_carla_agents.py 3/5 Renamed from process_carla_xodr.py, added argument parsing, AABB collision detection, initial velocity control, and trajectory generation with configurable parameters
data_utils/carla/plot.py 5/5 Added visualization script for plotting road data and agent trajectories from JSON files
data_utils/carla/JunctionOutlineTest.py 5/5 Added test/demo script for junction polygon creation with alpha shapes and visualization
external/pyxodr 2/5 Added git submodule pointer for pyxodr library (submodule appears empty in current checkout)

Sequence Diagram

sequenceDiagram
    participant User
    participant Git as Git Submodule System
    participant Script as generate_carla_agents.py
    participant PyXODR as pyxodr Library
    participant Network as RoadNetwork
    participant AABB as AABB Collision System
    participant JSON as Output JSON Files

    User->>Git: git submodule update --init --recursive
    Git->>PyXODR: Initialize pyxodr submodule
    
    User->>Script: python generate_carla_agents.py --args
    Script->>Script: Parse arguments (num_objects, avg_speed, etc.)
    
    loop For each town map
        Script->>PyXODR: Load XODR file
        PyXODR->>Network: Parse road network
        Network->>Script: Return roads, junctions, lane sections
        
        Script->>Script: Create lane link elements
        Script->>Script: Build successor/predecessor connections
        Script->>Script: Add incoming/outgoing edges
        
        loop For each agent (num_objects)
            Script->>Script: Select random lane and position
            Script->>AABB: Check collision with existing agents
            alt Collision detected
                Script->>Script: Retry with different position
            else No collision
                Script->>Script: Generate trajectory with velocity
                Script->>Script: Calculate waypoints along lane
                Script->>Script: Adjust Z-coordinate for geometry
                Script->>JSON: Add agent with position, heading, velocity
            end
        end
        
        Script->>JSON: Save Town_X_Y.json with agents
    end
    
    User->>Script: python data_utils/carla/plot.py
    Script->>JSON: Load and visualize agents on map
Loading

Copy link

@greptile-apps greptile-apps bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

21 files reviewed, 1 comment

Edit Code Review Agent Settings | Greptile

Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
@daphne-cornelisse daphne-cornelisse self-requested a review December 17, 2025 03:07
Copy link

@daphne-cornelisse daphne-cornelisse left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice work. Couple of comments:

  • Why do we need the pyxodr submodule? From the repo, it looks like you can just pip install it, in which case you can add it to the dependencies in setup.py
  • Thanks for adding docs. Can you please move it to https://emerge-lab.github.io/PufferDrive/data/ ? -> just add a section for Carla
  • Why do we have duplicate maps? Can't we just delete the maps under carla and keep the ones under carla_py123d?

@mpragnay
Copy link
Author

Nice work. Couple of comments:

  • Why do we need the pyxodr submodule? From the repo, it looks like you can just pip install it, in which case you can add it to the dependencies in setup.py
  • Thanks for adding docs. Can you please move it to https://emerge-lab.github.io/PufferDrive/data/ ? -> just add a section for Carla
  • Why do we have duplicate maps? Can't we just delete the maps under carla and keep the ones under carla_py123d?

The pyxodr submodule had issues related to map resolution, lane linkage and also the z coordinates. I worked on this part sometime ago and figured pyxodr wasn't really maintained(as my PRs raised to their repo weren't reviewed) so I decided we should continue on its fork.

Oh yea forgot to mention about the map duplication, the carla_py123d ones are the ones generated using py123d and the carla ones are those me and aditya cleaned post filtering with pyxodr, I decided to keep both just as a backup. Maybe I can create a separate branch with thos maps(Like backup carla maps or something). Yeah will add the docs there.

@mpragnay
Copy link
Author

Nice work. Couple of comments:

  • Why do we need the pyxodr submodule? From the repo, it looks like you can just pip install it, in which case you can add it to the dependencies in setup.py
  • Thanks for adding docs. Can you please move it to https://emerge-lab.github.io/PufferDrive/data/ ? -> just add a section for Carla
  • Why do we have duplicate maps? Can't we just delete the maps under carla and keep the ones under carla_py123d?

The pyxodr submodule had issues related to map resolution, lane linkage and also the z coordinates. I worked on this part sometime ago and figured pyxodr wasn't really maintained(as my PRs raised to their repo weren't reviewed) so I decided we should continue on its fork.

The carla_py123d ones are the ones generated using py123d and the carla ones are those me and aditya cleaned post filtering with pyxodr, I removed the extra ones from here and stored it in a separate branch. Added into docs as well.

@daphne-cornelisse
Copy link

The pyxodr submodule had issues related to map resolution, lane linkage and also the z coordinates. I worked on this part sometime ago and figured pyxodr wasn't really maintained(as my PRs raised to their repo weren't reviewed) so I decided we should continue on its fork.
The carla_py123d ones are the ones generated using py123d and the carla ones are those me and aditya cleaned post filtering with pyxodr, I removed the extra ones from here and stored it in a separate branch. Added into docs as well.

Makes sense! Thanks for cleaning up those files.

LGTM - last thing: Can you delete the docs from the readme now that they are under data/

@daphne-cornelisse daphne-cornelisse self-requested a review December 17, 2025 04:20
@mpragnay
Copy link
Author

The pyxodr submodule had issues related to map resolution, lane linkage and also the z coordinates. I worked on this part sometime ago and figured pyxodr wasn't really maintained(as my PRs raised to their repo weren't reviewed) so I decided we should continue on its fork.
The carla_py123d ones are the ones generated using py123d and the carla ones are those me and aditya cleaned post filtering with pyxodr, I removed the extra ones from here and stored it in a separate branch. Added into docs as well.

Makes sense! Thanks for cleaning up those files.

LGTM - last thing: Can you delete the docs from the readme now that they are under data/

Done!

@daphne-cornelisse daphne-cornelisse merged commit f349b24 into main Dec 17, 2025
14 checks passed
@daphne-cornelisse daphne-cornelisse deleted the Carla_Junction_Filter branch December 30, 2025 14:23
m2kulkarni added a commit to Emerge-Lab/Adaptive_Driving_Agent that referenced this pull request Feb 1, 2026
* Goal behavior fixes (Emerge-Lab#124)

* Make sure we can overwrite goal_behavior from python side and other minor improvements.

* Fix stop goal behavior bug.

* Make goal radius configurable for WOSAC eval.

* Reset to defaults + cleanup.

* Minor

* Minor

* Incorprate feedback.

* Update drive.h

Accel is being cut in half for no reason

* Add mode to only control the self-driving car (SDC) (Emerge-Lab#130)

* Add control mode.

* Fix error message.

* Fix incorrect obs dim in draw_agent_obs (Emerge-Lab#109)

* Fix incorrect obs dim in draw_agent_obs

* Update drive.h

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Replace product distribution action space with joint distribution (Emerge-Lab#104)

* make joint action space, currently uses multidiscrete and should be replaced with discrete

* Fix shape mismatch in logits.

* Minor

* Revert: Puffer doesn't like Discrete

* Minor

* Make action dim conditional on dynamics model.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Replace default ent_coef and learning_rate hparams (Emerge-Lab#134)

* Replace default learning rate and ent_coef.

* Minor

* Round.

* Add new weights binary with joint action space. (Emerge-Lab#136)

* Add support for logging optional evals during training (Emerge-Lab#133)

* Quick integration of WOSAC eval during training, will clean up tomorrow.

* Refactor eval code into separate util functions.

* Refactor code to support more eval modes.

* Add human replay evaluation mode.

* Address comments.

* Fix args and add to readme

* Improve and simplify code.

* Minor.

* Reset to default ini settings.

* Test for ini parsing (python and C) (Emerge-Lab#116)

* Add python test for ini file parsing

- Check values from default.ini
- Check values from drive.ini
- Additional checks for comments capabilities

* Add C test for ini file parsing

- Add CMake project to configure, build and test
- Test value parsing
- Test comments format
- Add comments for (un)expected results

* FIX: Solve all memory errors in tests

- Compile with asan

* Remove unprinted messages

* Add utest to the CI

- Ini parsing tests
- Update comments to clarify intent

* Update tests/ini_parser/ini_tester.c

- Change check conditions to if/else instead of ifs
- Speed up parsing speed (exist as soon as match is found)

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update tests/ini_parser/ini_tester.c

- Fix mismatch assignation

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* FIX: Move num_map to the high level of testing

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Fix missing arg (Emerge-Lab#141)

* Add WOSAC interaction + map metrics. Switch from np -> torch. (Emerge-Lab#138)

* Adding Interaction features

Notes:
- Need to add safeguards to load each map only once
- Might be slow if we increase num_agents per scenario, next step will
be torch.

I added some tests to see the distance and ttc computations are correct,
and metrics_sanity_check looks okay. I'll keep making some plots to
validate it.

* Added the additive smoothing logic for Bernoulli estimate.

Ref in original code:
message BernoulliEstimate {
    // Additive smoothing to apply to the underlying 2-bins histogram, to avoid
    // infinite values for empty bins.
    optional float additive_smoothing_pseudocount = 4 [default = 0.001];
  }

* Little cleanup of estimators.py

* Towards map-based realism metrics:

First step: extract the map from the vecenv

* Second step: Map features (signed distance to road edges)

A bunch of little tests in test_map_metric_features.py to ensure this do what it is supposed to do.

python -m pufferlib.ocean.benchmark.test_map_metrics

Next steps should be straightforward.

Will need to check at some point if doing this on numpy isnt too slow

* Map-based features.

This works, and passes all the tests, I would still want to make additionnal checks with the renderer because we never know.

With this, we have the whole set of WOSAC metrics (except for traffic lights), and we might also have the same issue as the original WOSAC code: it is slow.

Next step would be to transition from numpy to torch.

* Added a visual sanity check, plot random  trajectories and indicate when WOSAC sees an offorad or a collision

python pufferlib/ocean/benchmark/visual_sanity_check.py

* Update WOSAC control mode and ids.

* Eval mask for tracks_to_predict agents

* Replacing numpy by torch for the computation of interaction and map metrics.

It makes the computation way faster, and all the tests pass.

I didn't switch kinematics to torch because it was already fast, but I might make the change for consistency.

* Precommit

* Resolve small comments.

* More descriptive error message when going OOM.

---------

Co-authored-by: WaelDLZ <wawa@CRE1-W60060.vnet.valeo.com>
Co-authored-by: Waël Doulazmi <wawa@10-20-1-143.dynapool.wireless.nyu.edu>
Co-authored-by: Waël Doulazmi <wawa@Waels-MacBook-Air.local>
Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Multi map render support to wandb (Emerge-Lab#143)

Co-authored-by: Pragnay Mandavilli <pm3881@gr052.hpc.nyu.edu>

* Add mode for controlled experiments (Emerge-Lab#144)

* Add option for targeted experiments.

* Rename for clarity.

* Minor

* Remove tag

* Add to help message and make deepcopy of args to prevent state pollution.

* Little optimizations to use less memory in interaction_features.py (Emerge-Lab#146)

* Little optimizations to use less memory in interaction_features.py

They mostly consist in using in-place operations and deleting unused variables.

Code passes the tests.

Next steps:
- clean the .cpu().numpy() in ttc computation
- memory optimization for the map_features as well

* Add future todo.

---------

Co-authored-by: Waël Doulazmi <waeldoulazmi@gmail.com>

* Fix broken link

* Data processing script that works decent. (Emerge-Lab#150)

* Pass `map_dir` to the env via `.ini` and enable evaluation on a different dataset (Emerge-Lab#151)

* Support train/test split with datasets.

* Switch defaults.

* Minor.

* Typo.

* More robust way of parsing the path.

* Add sprites in headless rendering (Emerge-Lab#152)

* Load the sprites inside eval-gif()

* Color consistency.

* pedestrians and cyclists 3d models

* Minor.

---------

Co-authored-by: Spencer Cheng <spenccheng@gmail.com>

* Faster file processing (Emerge-Lab#153)

* multiprocessing and progbar

* cleanup

* Add link to small clean eval dataset

* Fix link typo

* Gif for readme (Emerge-Lab#155)

* Test

* Edit.

* Edit.

* Fix link?

* Fix vertical spaces.

* Update README.md

* Several small improvements for release (Emerge-Lab#159)

* Get rid of magic numbers in torch net.

* Stop recording agent view once agent reaches first got goal. Respawning vids look confusing.

* Add in missing models for headless rendering.

* Fix bbox rotation bug in render function.

* Remove magic numbers. Define constants once in drive.h and read from there.

* WIP changes (Emerge-Lab#156)

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Releas note

* Remove magic numbers in `drivenet.h`, set `MAX_AGENTS=32` by default (Emerge-Lab#165)

* Get rid of magic numbers in torch net.

* Stop recording agent view once agent reaches first got goal. Respawning vids look confusing.

* Add in missing models for headless rendering.

* Fix bbox rotation bug in render function.

* Remove magic numbers. Define constants once in drive.h and read from there.

* Remove all magic numbers in drivenet.h

* Clean up more magic numbers.

* Minor

* Minor.

* Stable: Ensure all tests are passing (Emerge-Lab#168)

* Test the tests

* Fix

* Add option to zoom in on the map or show full map (Emerge-Lab#163)

* Modifying render to view full map

* Removing vlue lines from maps

* Add option to zoom in on the map.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Add documentation (Emerge-Lab#170)

* Documentation first pass

* incorporate previous docs

* style updates and corrections

* Add GitHub Actions workflow for docs deployment (Emerge-Lab#172)

* styling fixes (Emerge-Lab#173)

* Add clang format (Emerge-Lab#132)

* Add clang format

- Format C/C++/CUDA
- Prevent formatting json
- Prevent formatting pufferlib extensions

* [FORMAT] Apply clang format

- No code changes

* Add clang format

- Format C/C++/CUDA
- Prevent formatting json
- Prevent formatting pufferlib extensions

* [FORMAT] Apply clang format

* Keep matrix printing as it is
* No code change

* small default change.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Add Sanity Command + Maps (Emerge-Lab#175)

* initial commit

* Ignore generated sanity binaries and compute ego speed along heading

* Fix get_global_agent_state to include length/width outputs

* Revert ego speed change in drive.h

* Add sanity runner wiring and wandb name hook

* Revert drive path changes; use map_dir from sanity

* Set sanity runs to use generated map_dir and render map

* Expand map path buffer in drive binding to avoid overflow

* fix maps and add docs

* update readme with documentation link

* Simplify docs.

* Apply precommit.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Documentation edits (Emerge-Lab#176)

* Softer theme and edits.

* Improve structure.

* Blog post v0

* Typo

* Fixes

* Early environment resets based on agents' respawn status.  (Emerge-Lab#167)

* Added early termination parameter based on respawn status of all agents in an episode

* pre-commit fix

* fix test

* Apply precommit.

* Reduce variance in aggregate metrics by logging only if we have data for at least num_agents.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Speed up end-to-end training: 220K -> 320K on RTX 4080 by reducing # road points (score maintained) (Emerge-Lab#177)

* 220K -> 320K.

* Reward bug fix.

* Minor.

* Add pt. (Emerge-Lab#179)

* Docs edits (Emerge-Lab#178)

* Simplify data downloading.

* Add links.

* Update WOSAC eval section.

* Minor.

* Rephrase.

* Fixes.

* Naming edits.

* Naming edits.

* There is a typo in torch.py

* Use num_maps for eval (Emerge-Lab#164)

* Use num_maps for eval

* readme.md didnt pass precommit ?

* Add the use_all_maps arg in the call to resample

* Update the wosac sanity checks to use all maps as well. Nicer prints in the eval

* Add a comment in drive.ini

* Update HumanReplay to follow the same logic

* Remove num_agents from WOSAC eval as it is not needed anymore.
Update the comments in drive.ini

* Change a comment in drive.py

* Update wosac.md

* Evaluated the base policy on validation interactive dataset and updated wosac.md with its score.

Also put back default behavior in drive.ini

* Fix small bug in `drive.c` and add binary weights cpt  (Emerge-Lab#184)

* Add model binary file and make demo() work.

* Add docs.

* Add docs.

---------

Co-authored-by: Daphne <daphn3cor@gmail.com>

* Carla junction filter (Emerge-Lab#187)

* Added Z Coords, Polygonized Junction Area to handle point in polygon query

* Added Snapped Polylines to better polygonizing

* Fixed Extra Road Lines bug, better polygonization with debugging

* Fixed initial heading angle for Trajectory sampling

* Maps Before manual filtering

* Carla Before Manual with z coordinates

* NaN fixes

* Minor

* Carla Maps Cleaned 6/8

* Add pyxodr submodule at external/pyxodr

* added external submodule support, fixed num_tries for valid init position of agents, added arg parsing, cleaned up code

* Removed unstable process_roads_script, use carla_py123d

* add avg_speed as arg

* Remove old Carla Maps

* Update README.md

Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>

* Remove old jsons and testing files

* Remove redundant instructions from README

* indentation changes

* Minor editorial edits.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Co-authored-by: Daphne <daphn3cor@gmail.com>

* Working Carla Maps (Emerge-Lab#189)

* collision fix (Emerge-Lab#192)

* collision fix

* lowered to 250 for actual theoretical guarantee

* Fix Ego Speed Calculation (Emerge-Lab#166)

* initial commit

* remove function

* restore weights, add trailing newline, fix ini

* update to unify logic and refactor to function

* fix reverse computation

* precommit fix

---------

Co-authored-by: Kevin <kj2676@nyu.edu>

* Small bug fix that makes road edge not appear in agent view for jerk model. (Emerge-Lab#197)

* add womd video (Emerge-Lab#195)

* Documentation first pass

* incorporate previous docs

* style updates and corrections

* initial commit

* Format drive helpers

* Add stop/remove collision behavior back (Emerge-Lab#169)

* Adding collision behavior back

* Removing uneccesary snippets

* Rebased

* precommit fixes

* Pre-Commit fixes

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* updated docs with multinode training cmd (Emerge-Lab#174)

* Carla2d towns (Emerge-Lab#201)

* Added valid 2d carla towns, some code cleanup

* Add carla 3D maps with objects

* initial commit (Emerge-Lab#204)

* Fix goal resampling in Carla maps and make metrics suitable for resampling in longer episodes (Emerge-Lab#186)

* Make road lines and lanes visible in map.

* Simplify goal resample algorithm: Pick best road lane point in road graph.

* Delete redundant code.

* Make the target distance to the new goal configurable.

* Generalize metrics to work for longer episodes with resampling. Also delete a bunch of unused graph topology code.

* Minor

* Apply precommit.

* Fix in visualizer.

* fix metrics

* WIP

* Add goal behavior flag.

* Add fallback for goal resampling and cleanup.

* Make goal radius more visible.

* Minor

* Make grid appear in the background.

* Minor.

* Merge

* Fix bug in logging num goals reached and sampled.

* Add goal taret

* Use classic dynamics model.

* Fix descrepancies between demo() and eval_gif().

* Small bug fix.

* Reward shaping

* Termination mode must be 0 for Carla maps.

* Add all args from ini to demo() env.

* Clean up visualization code.

* Clean up metrics and vis.

* Fix metrics.

* Add diversity to agent view.

* Add better fallback.

* Reserve red for cars that are in collision.

* Keep track of current goals.

* Carla testing simple/

* Use classic dynamics by default.

* Fix small bug in goal logging (respawn).

* Always draw agent obs when resampling goals.

* Increase render videos timeout (carla maps take longer).

* Minor vis changes.

* Minor vis changes.

* Rmv displacement error for now and add goal speed target.

* Add optional goal speed.

* Incorporate suggestions.

* Revert settings.

* Revert settings.

* Revert settings.

* Fixes

* Add docs

* Minor

* Make grid appear in background.

* Edits.

* Typo.

* Minor visual adaptations.

---------

Co-authored-by: Daphne <daphn3cor@gmail.com>
Co-authored-by: julianh65 <jhunt17159@gmail.com>

* Minor correction in resampling code (Emerge-Lab#183)

* Corrections of the resample code in drive.py:

- the will_resample=1 followed by if will_resample looked weird to me (probably legacy code ?)
- When we resample we should update the values of self.agent_offsets map dirs and num envs.

The fact that we didn't update them isn't an issue because right now they are not accessed anywhere in the code, but then we should either remove these attributes of the Drive Class or either make ensure they contain the right values if someone wants to use them later.

* Minor

* Fix merge conflicts.

---------

Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Allow human to drive with agents through classic and jerk dynamics model (Emerge-Lab#206)

* Fix human control with joint action space & classic model: Was still assuming multi-discrete.

* Enable human control with jerks dynamics model.

* Color actions yellow when controlling.

* Slightly easier control problem?

* Add tiny jerk penalty: Results in smooth behavior.

* Pre-commit

* Minor edits.

* Revert ini changes.

---------

Co-authored-by: Daphne <daphn3cor@gmail.com>

* Added WOSAC results on the 10k validation dataset (Emerge-Lab#185)

* Added WOSAC results on the 10k validation dataset

* Code to evaluate SMART + associated doc

* Edits.

* Add link to docs.

---------

Co-authored-by: Wael Boumediene Doulazmi <wbd2016@gl001.hpc.nyu.edu>
Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>

* Drive with agents in browser (Emerge-Lab#215)

* Good behavior with trained policy - resampling.

* Hardcode to 10 max agents for web version.

* Browser demo v1.

* More descriptive docs.

* Release post edits.

* Docs improvements.

* Run precommit.

* Better policy.

* Revert .ini changes, except one.

* Delete drive.dSYM/Contents/Info.plist

* Delete pufferlib/resources/drive/puffer_drive_gljhhrl6.bin

---------

Co-authored-by: Daphne <daphn3cor@gmail.com>

* Fix demo (Emerge-Lab#217)

Co-authored-by: Daphne <daphn3cor@gmail.com>

* Do not randomly switch to another agent in FPV. (Emerge-Lab#219)

Co-authored-by: Daphne <daphn3cor@gmail.com>

* switch docs to mdbooks doc format (Emerge-Lab#218)

Move from mkdocs to mdbooks. Code heavily claude assisted.

* Markdown edits and fix demo. (Emerge-Lab#221)

Co-authored-by: Daphne <daphn3cor@gmail.com>

* small fixes in the docs (Emerge-Lab#220)

* fix minor errors

* try to fix things for dark mode

* Fixing dark/light mode error

---------

Co-authored-by: Eugene Vinitsky <eugene@percepta.ai>
Co-authored-by: Aditya Gupta <adigupta2602@gmail.com>

* Release 2.0 (Emerge-Lab#214)

* Ensure the arrows are on the ground.

* Date fix.

* Update data docs with mixed dataset.

* Increase map range.

* Remove ini file for web demo.

* WIP

* Edit release post.

* Minor docs fixes.

* Writing changes.

* Fix metrics.

* Delete outdated readme files.

* Minor.

* Improve docs.

* Fix institutions.

* Update training info.

* Reset configs to womd defaults.

* Update score metrics.

* Update docs

* Update policy for demo.

* Update demo files (new cpt)

* Minor.

* Add video.

* Minor.

* Keep defaults.

---------

Co-authored-by: Daphne <daphn3cor@gmail.com>

* Fix space and game files.

* Fix sup tags.

* Self play working

* Population play and self play rebased

* All features working

* fixing co player features

* trying to pass tests

* fixing tests #2

* fixing tests  #3

* attempting to fix tests #4

* attempting to fix tests #4

* attempting to fix tests #4

* fixing batch size > 1 bug

* add back binary

* changed map dir

* moved maps

* fix test API and config

* fixing tests

* fix tests

---------

Co-authored-by: Daphne Cornelisse <33460159+daphne-cornelisse@users.noreply.github.com>
Co-authored-by: Eugene Vinitsky <eugenevinitsky@users.noreply.github.com>
Co-authored-by: Daphne Cornelisse <cor.daphne@gmail.com>
Co-authored-by: AJE <231052006+aje-valeo@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Waël Doulazmi <73849155+WaelDLZ@users.noreply.github.com>
Co-authored-by: WaelDLZ <wawa@CRE1-W60060.vnet.valeo.com>
Co-authored-by: Waël Doulazmi <wawa@10-20-1-143.dynapool.wireless.nyu.edu>
Co-authored-by: Waël Doulazmi <wawa@Waels-MacBook-Air.local>
Co-authored-by: Pragnay Mandavilli <108453901+mpragnay@users.noreply.github.com>
Co-authored-by: Pragnay Mandavilli <pm3881@gr052.hpc.nyu.edu>
Co-authored-by: Waël Doulazmi <waeldoulazmi@gmail.com>
Co-authored-by: Spencer Cheng <spenccheng@gmail.com>
Co-authored-by: Kevin Joseph <kevinwinston184@gmail.com>
Co-authored-by: Aditya Gupta <adigupta2602@gmail.com>
Co-authored-by: Julian Hunt <46860985+julianh65@users.noreply.github.com>
Co-authored-by: riccardosavorgnan <22272744+riccardosavorgnan@users.noreply.github.com>
Co-authored-by: Daphne <daphn3cor@gmail.com>
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Co-authored-by: Kevin <kj2676@nyu.edu>
Co-authored-by: julianh65 <jhunt17159@gmail.com>
Co-authored-by: Wael Boumediene Doulazmi <wbd2016@gl001.hpc.nyu.edu>
Co-authored-by: Eugene Vinitsky <eugene@percepta.ai>
Co-authored-by: charliemolony59@gmail.com <cpm9831@ga021.hpc.nyu.edu>
Co-authored-by: charliemolony59@gmail.com <cpm9831@ga009.hpc.nyu.edu>
Co-authored-by: charliemolony59@gmail.com <cpm9831@ga014.hpc.nyu.edu>
Co-authored-by: charliemolony59@gmail.com <cpm9831@ga019.hpc.nyu.edu>
Co-authored-by: Mohit Kulkarni <mkulkarni@ethz.ch>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants