Deadline Extended to Feb 8 + Instructions for Final Submission Package

there is no @aicrowd in the gitlab.

image508×604 49.2 KB

View post 0

Duplicate failed submissions, requesting for a final submission slot

I am just curious.

Our code should be the same with what we submitted for the evaluation test? (with the best score)

Our code will be re-executed for the hidden test-case.

Then, we don’t need to actually submit for evaluation.

Leather board score is a reference, which is not the final score.

View post 0

Deadline Extended to Feb 8 + Instructions for Final Submission Package

For Track 2 (who might use commercialized APIs without fine-tuned),

can we skip to submit Model artifacts and documentation, Reproducibility artifacts?

calling OpenAI API is all

View post 0

Missing processed_map directory causes remote evaluation failure

I think something wrong in the recent submission server.

• “Confirming if the processed_map files should be part of the official starter-kit or if they should be uploaded manually by participants.”
  → I don’t think so, it worked automatically in the previous version.

And there is also module load error in the recent start-kit version:
ModuleNotFoundError: No module named ‘evaluation_utils.live_export’ - Orak Game Agent Challenge - AIcrowd Forum

No one success the submission in one-week.

image1930×1788 318 KB

View post 1

Pokemon, does the high-level-tools work?

the recent startkits have that use_tool function.

look here.

View post 0

ModuleNotFoundError: No module named 'evaluation_utils.live_export'

In the recent starterkit,

• evaluation_utils/live_export.py file is missing.

File “C:\MyProjectFolders\orak-2025-starter-kit-master_new\evaluation_utils\game_server_launcher.py”, line 10, in from evaluation_utils.live_export import live_export_enabled

ModuleNotFoundError: No module named ‘evaluation_utils.live_export’

View post 1

Pokemon, does the high-level-tools work?

I was using old code.
The recent code in git have that function.

View post 0

Pokemon, does the high-level-tools work?

There is high-level tools in poketmon based on the doc.

(1) Does it work? (i failed…, though simple command worked such as A,B button)

• return ‘use_tool(move_to, (x=7 , y=1))’

(2) Using the high-level tool is counted as 1 step?

if the high-level tool is counted as a 1 step, we have to use it to reduce the steps. (200 steps is very limited). Then, I can’t continue to work on pokemon without the tools.

Anyone can successfully use the high-level tools in poketmon?

• move_to(x, y): Finds and executes a path to the specified map coordinates (x,y).
• interact_with_object(name): Interacts with a specified object in the environment.
• warp_with_warp_point(x, y): Uses a specified warp point to move to a different location.
• overworld map transition(direction): Transitions to an adjacent map in the given direction.
• continue_dialog(): Advances the current dialogue.
• select_move_in_battle(move_name): Selects and uses a specific move in a Pokemon battle.
• switch_pkmn_in_battle(pokemon_name): Switches to a different Pokemon in the player’s party during a battle.
• run_away(): Attempts to flee from a wild Pokemon battle.
• use_item_in_battle(item_name): Uses a specified item during a Pokemon battle.

Example:

use_tool(select_move_in_battle, (move_name="Thunder Shock"))

View post 0

Pokemon map is broken?

Map broken issue is fixed after case-insensitivy update.

View post 0

Star2 Crash Issue

• STAR2 MAX_STEPS issue

MAX_STEPS value doesn’t work in Star2 in original code.

I have a game log that played 2 hours.
there were 2 reasons
(1) ‘MULTI-ATTACK’ cannot destroy all of enemy’s building (some of theirs are built in the corner)
(2) MAX_STEPS (ex., 1000 steps) didn’t work.

in star_craft_env.py.

def check_process(self, reset=False):
    ctx = getattr(self, "_mp_ctx", multiprocessing.get_context())
    if self.p is not None:
        if self.p.is_alive():            
            #if not self.game_over.value:  # should be changed like below. after MAX_STEPS doesn't mean it is game_over
            if not self.game_over.value and not reset:  # Check if the game is over
                return  # Return ONLY if game is running AND we are NOT resetting (MAX_STEP)

            # KILL PROCESS LOGIC (depends on the OS)          
            self.p.terminate()

To re-make this bug,
in common.py
change star_craft value to low, it is not gonna finished after that max_step.

MAX_STEPS = {
“twenty_fourty_eight”: 1000,
“super_mario”: 100,
“pokemon_red”: 200,
“star_craft”: 50,
}

View post 0

Star2 Crash Issue

Thanks for your response.

I’ve identified several issues, and I’ll describe them starting with the most critical one.

(1) Crash

This is a critical crash that causes the simulation to fail and therefore must be fixed.
After applying the fix, building actions can still fail.

**(2) Multi-RETREAT **

When enemy is going back to their base (after fight), my units automatically follow them to the enemy main base.

In this situation, it appears that the retreat target position is updated to the combat destination (i.e., the enemy main base), causing the retreat behavior to effectively function the same as an attack command.

As a result, there is currently no reliable way to make units retreat.

image1920×1163 239 KB

(3) Minimum distance for building placement

There seems to be a minimum distance constraint when placing buildings such as pylons.
On extremely small maps like FLAT64, this can result in pylons being placed very close to the enemy main base.

image922×620 57.5 KB

(4) Replay file (minor)
Each execution runs three episodes. However, the replay files use the same filename and overwrite each other, so only one replay file remains.
If replays are intended to be saved per episode, the filename logic would need to be modified to generate a separate replay file for each episode.

While it is true that participants are expected to account for these constraints, this is quite unexpected situation when playing startcraft.
and I understand that some participants have already developed under this environment, so too late to
update the code.

View post 1

Pokemon map is broken?

so… did you mean that
the current evaluation server has the map loading issue (such as /? /?).
Then, the score in the leather board was achevied without the map information…?

I haven’t tried poketmon for this contest because of the map loading problem.
I will try your solution.

Thanks in advance.

View post 0

Star2 Crash Issue

This map is not good.
My units can’t go attack because of the ROCK obstacle. (They stucked there almost a few minutes after ‘MULTI-ATTACK’)

View post 0

Star2 Crash Issue

I hate this simcity

View post 0

Star2 Crash Issue

This is very critical…
When they just skip to build an important building such as PYLON.

It’s very confusing should I consider, this will be happened in real-test too, so should I try to over-building.

If someone wants to re-make this bug, just keep building ‘BUILD PYLON’, after three of them it just stop building.

View post 0

Obs_str position misalignment: Mario's coordinates gradually drift from the actual position

Possible Mario simulation bug? - Orak Game Agent Challenge - AIcrowd Forum

The same issue (gradual misalignment) with me.

I think this is kinda of bug(?) from the simulator.
I don’t know how to fix it.

When I see the screenshots, the mario is gradually going forward but the position_x of text is the same.

So, I just think that to get 1.0 point, we have to pass the monters through upstrair-blocks (diffrent y_axis with the mosters)

View post 1

Star2 Crash Issue

so, defending with cannons is not gonna work.
many fails happen when try to build cannons.
(couldn’t find best_position)

View post 1

Star2 Crash Issue

There is a crash issue in StarCraft 2.

While executing a strategy of defending with Photon Cannons followed by an attack, the simulation would irregularly crash and terminate.

Upon debugging, the problem was traced to building placement. According to bots.py, a building can only be constructed once best_position is determined. When the surrounding area search repeatedly fails to find a suitable location, best_position remains None, causing a crash when attempting to build.

In the bots.py file, within the function:

python

async def handle_action_build_building(self, building_type: UnitTypeId, building_limits=None):

There is this commented section:

python

# if not best_position:
#     print(f"Still no suitable position found for {building_type}. Aborting.")
#     return

Fix: Uncomment this section so that if best_position cannot be found, the function simply skips the build:

python

if not best_position:
    print(f"Still no suitable position found for {building_type}. Aborting.")
    return

This change prevents the crash by gracefully handling cases where no valid building location is found, allowing the bot to continue running instead of terminating the simulation.​

View post 1

Impossible to run Star2 in Linux (no map file)

(1)
there is no the map file in the links. Ancient Cistern LIE
Maps - AI Arena Wiki
I’ve downloaded all the file, but there isn’t.
Also, some links are broken in the url.

(2)
GitHub - aiarena/sc2patch

it’s difficult to find the instructions in the link.
Maybe, there is youtube link to apply the patch.
They used SC2 map editor? and updated config file?
it’s quite difficult to understand what they are doing for me.

It would be easy to simply share the map files for this contest for the linux users I believe.

View post 0

Impossible to run Star2 in Linux (no map file)

• I spent a few hours trying to run the Star2 simulation on Linux, but I couldn’t get it to work. The biggest issue was that the “Ancient Cistern LIE” file could not be found even in the whole internet.

• So, I reconfigured the ORAK simulation on Windows instead. There was a gRPC-related issue, but it was resolved by modifying game_server_launcher.py as shown below:
  cmd = [
  sys.executable, # fixed
  game_server_script,
  ]
  env = os.environ.copy()

• On Windows, I renamed the “Ancient Cistern LIE” file to “Ancient Cistern LE” and downloaded the file into /Maps/. When running the simulation, I could clearly see the game launch and play screen, which makes prompt debugging much easier.
  On Linux, I was not able to see the StarCraft play screen during execution.

View post 0