metadata

license: apache-2.0
task_categories:
  - robotics

Dataset

This dataset is used to train a transporter network for real-world pick/place within the RAD lab at the University of Edinburgh. The dataset is in TFDS format and was collected using moveit2_data_collector. In its current state the dataset is being tested as we are proving out this overall pipeline, keep monitoring this dataset and related repos for documentation updates.

Download

An example of downloading and loading the dataset is given below, as larger datasets are uploaded this example script will change:

import os
import tarfile

import tensorflow_datasets as tfds
from huggingface_hub import hf_hub_download

DATA_DIR="/home/robot"
FILENAME="data.tar.xz"
EXTRACTED_FILENAME="data"
FILEPATH=os.path.join(DATA_DIR, FILENAME)
EXTRACTED_FILEPATH=os.path.join(DATA_DIR, EXTRACTED_FILENAME)

# download data from huggingface
hf_hub_download(
        repo_id="peterdavidfagan/transporter_networks", 
        repo_type="dataset",
        filename=FILENAME,
        local_dir=DATA_DIR,
        )

# uncompress file
with tarfile.open(FILEPATH, 'r:xz') as tar:
    tar.extractall(path=DATA_DIR)
os.remove(FILEPATH)

# load with tfds
ds = tfds.builder_from_directory(EXTRACTED_FILEPATH).as_dataset()['train']

# basic inspection of data
print(ds.element_spec)
for eps in ds:
    print(eps["extrinsics"])
    for step in eps["steps"]:
        print(step["is_first"])
        print(step["is_last"])
        print(step["is_terminal"])
        print(step["action"])