johannah

johannah.github.io

Blog powered by nikola

opendrift-predict

pytorch-nerf

Minimal PyTorch implementations of NeRF and pixelNeRF.

DH

Experiments for using a differentiable Denavit-Hartenberg loss to improve the speed and stability of learning robot control policies. New work on identifying DH parameters from experience. Ongoing research.

robomimic

robomimic: A Modular Framework for Robot Learning from Demonstration. Modifications for enabling efficient policy transfer.

balloon-learning-environment

The Balloon Learning Environment - flying stratospheric balloons with deep reinforcement learning.

CLIPasso

dexmv-sim

DexMV: Imitation Learning for Dexterous Manipulation from Human Videos, ECCV 2022

dissipative_hnns

Training neural networks to disentangle conservative and dissipative dynamics

drift-predict

drqv2

DrQ-v2: Improved Data-Augmented Reinforcement Learning

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

ffcv

FFCV: Fast Forward Computer Vision

frankapy

Python interface to control Franka Emika Panda Research Robot Arms.

gpu-io

A GPU-accelerated computing library for physics simulations and other mathematical calculations

gym-fish

handover-sim

A simulation environment and benchmark for human-to-robot object handovers

kinematic-critic

latent-diffusion

High-Resolution Image Synthesis with Latent Diffusion Models

mae

PyTorch implementation of MAE https//arxiv.org/abs/2111.06377

opendrift

Open source framework for ocean trajectory modelling

pybullet_utils

Simple utilities for PyBullet, including collision detection, ghost (i.e. visual-only) objects, and cameras.

rob501

Mathematics for Robotics

robosuite

robosuite: A Modular Simulation Framework and Benchmark for Robot Learning

ros_interface

SGSDI

tech-resume

umpnet

[RA-L / ICRA 2022] UMPNet: Universal Manipulation Policy Network for Articulated Objects

VQ-Diffusion

Official implementation of VQ-Diffusion

worldstrat

The WorldStrat Dataset