Location
Badges
Activity
Ratings Progression
Challenge Categories
Challenges Entered
Evaluate Natural Conversations
Latest submissions
Small Object Detection and Classification
Latest submissions
Understand semantic segmentation and monocular depth estimation from downward-facing drone images
Latest submissions
Audio Source Separation using AI
Latest submissions
See Allgraded | 209116 |
Identify user photos in the marketplace
Latest submissions
Using AI For Buildingβs Energy Management
Latest submissions
Interactive embodied agents for Human-AI collaboration
Latest submissions
Behavioral Representation Learning from Animal Poses.
Latest submissions
ASCII-rendered single-player dungeon crawl game
Latest submissions
Measure sample efficiency and generalization in reinforcement learning using procedurally generated environments
Latest submissions
See Allgraded | 94599 | ||
graded | 94551 | ||
graded | 93732 |
3D Seismic Image Interpretation by Machine Learning
Latest submissions
Multi-agent RL in game environment. Train your Derklings, creatures with a neural network brain, to fight for you!
Latest submissions
5 Problems 15 Days. Can you solve it all?
Latest submissions
Latest submissions
See Allgraded | 200973 | ||
graded | 200972 |
Participant | Rating |
---|---|
priteshgohil | 0 |
nachiket_dev_me18b017 | 0 |
cadabullos | 0 |
Participant | Rating |
---|
Data Purchasing Challenge 2022
IMPORTANT: Details about end of competition evaluations π―
Over 2 years ago@ArtemVoronov Its intentional. Please select 4.
NeurIPS 2021 AWS DeepRacer AI Driving Olympics Cha
Ideas for better sim2real transfer
Almost 3 years agoThis was your last successful submission on 24th November which is already updated since a few days. Failed submissions are not run on the real track. But seems like the your submission was consistently turning towards the right side so the score is lower than your previous submission.
Please check all your other successful submissions
after 20th November and let me know in case any are missing, we were running the best submission per user per day until recently. Weβll run any older ones in case any are missing.
Please let me know if I can help with anything else, Iβll try my best but might be bit delayed since Iβm travelling today.
Regarding evaluation criteria for online submissions
Almost 3 years agoSorry I missed reply to this this post earlier.
The reward function for both the rounds is staying close to the center, this is a proxy reward but its good enough as the action space does not allow stopping. The score on the simulation leaderboard is indeed the mean cumulative reward over multiple episodes. The maximum possible reward in both rounds may vary as the tracks are different.
But the main idea of the competition is the sim2real transfer and the formula for the real track is based on waypoints crossed and time. The final prizes are based on the real track leaderboard and a small weightage on the simulation scores. Iβll make a separate post to clarify the formula for the real track.
Round 2 is live!
About 3 years agoHi Everyone,
The much awaited Round 2 of the Deepracer Challenge is now live. Submissions will now be run on a real deepracer track everyday until November 30th.
The submission flow will be a follows:
Each team is allowed upto 5 successful submissions everyday, these will run on the Round 2 simulator.
The top scoring submission from each team each day will be run on the real track.
Since the time remaining for Round 2 is short, we encourage everyone to train their models on the Round 2 simulator as soon as possible.
There will be no obstacles on the track, but sim2real transfer will not be a trivial task. Be on the lookout for on board videos from your submissions to help you fine tune your solutions. All the best!!
β° Round 1 extended till Nov. 10th!
About 3 years agoWeβre in the process of setting up the track, thanks for the patience.
Updates for starter kit
About 3 years agoHi @asche_thor
This used to be an older issue but was fixed, are you on the latest version of the env docker release
Can you please pull the latest docker and check once.
Flatland
Some questions ahout the agents' speed
About 3 years agoSeems Iβd missed answering the question on varying speeds, thanks @sa7show for the answer. Indeed youβre right the speeds for each train are decided at the start of the episode and stays fixed throughout the episode.
NeurIPS 2021 - The NetHack Challenge
How to revise βSelect Submissionsβ form?
About 3 years ago@fxofxo You can make 3 new submissions in round 2 all of which will be considered for final evaluation, if this is not sufficient let us know we can arrange for changing old ones.
Example of using non-neural model
Over 3 years agoHi @pajansen
The env returns the observations that are already separated and encoded into a dictionary for convenience, this has nothing to do with neural net. Is there a particular format you need?
Colab Notebook including submission and working ttyrec recording
Over 3 years agoHi @charel_van_hoof, could you try once with git bash on Windows. Iβll try to check but I think nle is currently not support on windows. But just the ttyplay2 script might work.
IITM RL Final Project-b5d2e6
Submission Error : Inference Failed
Over 3 years agoSeems like the error is due to RAM usage of your code. The evaluations run with a limit of 4 GB RAM, please make sure to run within this limit.
Some garbage collection code in python might help.
Submission limit
Over 3 years agoYou should be able to test everything locally then submit.
Can you please share why the extra submissions are needed?
Episode Count in Runner
Over 3 years agoBSuite gives the limit of 10000 as a fair amount episodes needed to converge.
Generally speaking, to have a level playing field among all competitors, while also not making it too easy, some constraint has to he applied. In this case, the number of episodes serves that role.
I know the number 10000 may be arbitrary, so if enough students feel it should be increased, weβll do it. For now please try to improve your algorithm to get the highest score with 10000 episodes.
Best Regards,
Dipam
IIT-M RL-ASSIGNMENT-2-GRIDWORLD
Welcome to your 2nd Assignment!
Over 3 years agoUnfortunately the choice of giving extra test cases is not upto me, please talk to the TAs.
In my opinion the test cases provided are sufficient, please review your code properly.
Welcome to your 2nd Assignment!
Over 3 years agoCan you check if your local scoring cell is working without any error.
In case your local scoring cell gives an error its probably because the output format is wrong, check the targets file for the output format.
If your local scoring cell gives no error please let me know.
Welcome to your 2nd Assignment!
Over 3 years agoSounds like a formatting issue with the code on your end, since its not a general issue that affects all students I encourage you to find the bug on your own. With correct format you should get decimals for all algorithms in the local scoring code. Look at the targets for example of the format.
Do let me know in case the problem still persists after youβve checked it thoroughly.
Welcome to your 2nd Assignment!
Over 3 years agoIt looks like the error says you need to accept the challenge rules. Can you please accept the rules then try submitting again.
IIT-M RL-ASSIGNMENT-2-TAXI
Unable to submit post 20th april
Over 3 years agoWe recently fixed this, can you close the notebook and run again from start, that should fix it.
Multi-Agent Behavior: Representation, Model-17508f
Can we use training data from task 1 or 2 for task 3?
Over 3 years agoHi @sungbinchoi
Yes, you can use training data of Task 1 for 2 and 3. Feel free to use all the data at your disposal.
You can even use some unsupervised learning on the test sequences.
Notebooks
-
[Explainer] - EDA of Seismic data by geographic axis Hereβs my EDA notebook on how the seismic data varies by geographic axis, along with some ideas for training.dipam_chakrabortyΒ· About 4 years ago
IMPORTANT: Details about end of competition evaluations π―
Over 2 years agoHi @chuifeng
There will be no live submissions with these changes. You will have to select 2 of your submissions from Round 2 for the final evaluations. Form and instructions selecting submissions will be provided before the competition ends.