README.md

Downloading Datasets

This script works for both MuSoHu and SCAND. Download MuSoHu and SCAND and put bag files in separate directories. Change bags_dir in config file accordingly.

Dependencies :

To install dependencies for parser:

pip install omegaconf bagpy tqdm

Parsing

To run the MuSoHu parser:

python parser.py --name musohu

And to run SCAND parser, change the dataset --name to scand:

python parser.py --name scand

We only store the front facing camera for the Spot in SCAND, so both MuSoHu and SCAND have the same interface. The only difference is that SCAND does not contain depth data.

Additionally, you can pass a custom config file:

python parser.py -c conf/custom_config --name musohu

Creating Samples

To create samples from parsed bags, run the following command:

python parser.py -cs

The sampler uses the observation length (obs_len) and prediction length (pred_len) from the config file to create samples. It reads the directory for storing the pickle file from parsed_dir.