Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

How to run objectnav baseline?

See original GitHub issue

Hi all,

I’ve updated latest version of both habitat-api and habitat-sim and try to train object-nav baseline by simply running the code as follows: python -u habitat_baselines/run.py --exp-config habitat_baselines/config/objectnav/ddppo_objectnav.yaml --run-type train

However, after showing config parameters, the programme seems to stuck in this place: Initializing dataset ObjectNav-v1

I’ve properly installed mp3d and its dataset, but I run on a single gpu and I don’t have NCCL. Is that the cause? And what is the right procedure to run objectnav baseline?

Thank you!

Issue Analytics

State:
Created 4 years ago
Comments:5 (1 by maintainers)

Top GitHub Comments

1reaction

mathfaccommented, May 19, 2020

@qianLyu BrokenPipeError: [Errno 32] Broken pipe is usually cause by other error with environment running as separate processes. Try to check memory occupation on the GPU as well as decrease number of processes. MP3D 3D scenes take more memory than Gibson’s ones.

0reactions

rpartseycommented, Sep 1, 2022

Closing the issue because it has not had recent activity.

Top Results From Across the Web

How to run objectnav baseline? · Issue #304 - GitHub

I've updated latest version of both habitat-api and habitat-sim and try to train object-nav baseline by simply running the code as follows:

ObjectNav baselines - AllenAct

Training a baseline#. To train a baseline ResNet->GRU model taking RGB-D inputs, run the following command.

Getting Started — RoboTHOR ObjectNav Challenge ...

We use a model-agnostic evaluation process to measure the performance of your trained agent. Submissions are ranked on the leaderboard by metric Test...

Stubborn: A Strong Baseline for Indoor Object Navigation - arXiv

We present a strong baseline that surpasses the performance of previously published methods on the Habitat Challenge task of navigating to a ...

BEyond observation: an approach for ObjectNav

ing algorithms can perform the Embodied Artificial Intelli- ... task, a.k.a Object-Goal Navigation (ObjectNav) consists ... PPO) is the baseline.