PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav

Code for our paper PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav.

Ram Ramrakhya, Dhruv Batra, Erik Wijmans, Abhishek Das

Project Page

What is PIRLNav?

PIRLNav is a two-stage learning scheme for IL pretraining on human demonstrations followed by RL-finetuning for ObjectNav. To enable successful RL finetuning, we present a two-stage learning scheme involving a critic-only learning phase first that gradually transitions over to training both the actor and critic.

Scaling laws of IL→RL for ObjectNav

Using this IL→RL training recipe, we present a rigorous empirical analysis of design choices. We study how RL-finetuning performance scales with the size of the IL pretraining dataset. We find that as we increase the size of the IL-pretraining dataset and get to high IL accuracies, the improvements from RL-finetuning are smaller, and that 90% of the performance of our best IL→RL policy can be achieved with less than half the number of IL demonstrations.

Read more in the paper.

Installation

Run the following commands:

git clone https://github.com/Ram81/pirlnav.git
git submodule update --init

conda create -n pirlnav python=3.7 cmake=3.14.0

cd habitat-sim/
pip install -r requirements.txt
./build.sh --headless

pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 torchaudio==0.12.1 --extra-index-url https://download.pytorch.org/whl/cu113

cd habitat-lab/
pip install -r requirements.txt

pip install -e habitat-lab
pip install -e habitat-baselines

pip install -e .

Data

Downloading HM3D Scene and Episode Dataset

• Download the HM3D dataset using the instructions here (download the full HM3D dataset for use with habitat)

• Move the HM3D scene dataset or create a symlink at data/scene_datasets/hm3d.

• Download the ObjectNav HM3D episode dataset from here.

Download Demonstrations Dataset

You can use the following command to download the datasets to reproduce results reported in our paper.

git clone git@hf.co:datasets/axel81/pirlnav data/datasets/objectnav

Dataset	Scene dataset	Split	Link	Dataset path
ObjectNav-HD	HM3D	77k	objectnav_hm3d_hd	data/datasets/objectnav/objectnav_hm3d_hd/
ObjectNav-SP	HM3D	240k	objectnav_hm3d_sp	data/datasets/objectnav/objectnav_hm3d_sp/
ObjectNav-FE	HM3D	70k	objectnav_hm3d_fe	data/datasets/objectnav/objectnav_hm3d_fe/

The demonstration datasets released as part of this project are licensed under a Creative Commons Attribution-NonCommercial 4.0 License.

OVRL Encoder Weights

To train policies using OVRL pretrained RGB encoder, download the model weights from here and move to data/visual_encoders/.

Dataset Folder Structure

The code requires the datasets in data folder in the following format:

├── habitat-web-baselines/
│  ├── data
│  │  ├── scene_datasets/
│  │  │  ├── hm3d/
│  │  │  │  ├── JeFG25nYj2p.glb
│  │  │  │  └── JeFG25nYj2p.navmesh
│  │  ├── datasets
│  │  │  ├── objectnav/
│  │  │  │  ├── objectnav_hm3d/
│  │  │  │  │  ├── objectnav_hm3d_hd/
│  │  │  │  │  │   ├── train/
│  │  │  │  │  ├── objectnav_hm3d_v1/
│  │  │  │  │  │   ├── train/
│  │  │  │  │  │   ├── val/
│  │  ├── visual_encoders

Usage

IL Training

For training the behavior cloning policy on the ObjectGoal Navigation task use the following script:

sbatch scripts/1-objectnav-il.sh <dataset_name>

where dataset_name can be objectnav_hm3d_hd, objectnav_hm3d_sp, or objectnav_hm3d_fe

RL Finetuning

For RL finetuning the behavior cloned policy on the ObjectGoal Navigation task use the following script:

sbatch scripts/2-objectnav-rl-ft.sh /path/to/initial/checkpoint

Evaluation

To evaluate a checkpoint trained using behavior cloning checkpoint use the following command:

sbatch scripts/1-objectnav-il-eval.sh /path/to/checkpoint

For evaluating a checkpoint trained using RL finetuning use the following command:

sbatch scripts/1-objectnav-rl-ft-eval.sh /path/to/checkpoint

Reproducing Results

We provide best checkpoints for agents trained on ObjectNav task with imitation learning and RL finetuning. You can use the following checkpoints to reproduce results reported in our paper.

Task	Checkpoint	Success Rate	SPL
🆕ObjectNav	objectnav_il_hd.ckpt	64.1	27.1
🆕ObjectNav	objectnav_rl_ft_hd.ckpt	70.4	34.1

Citation

If you use this code in your research, please consider citing:

@inproceedings{ramrakhya2023pirlnav,
      title={PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav},
      author={Ram Ramrakhya and Dhruv Batra and Erik Wijmans and Abhishek Das},
      booktitle={CVPR},
      year={2023},
}