pyRad 🤘

An all-in-one repo for NeRFs

Quickstart

Installation: Setup the environment

# Clone the repo
git clone --recurse-submodules git@github.com:ethanweber/pyrad.git

# Create the python environment
conda create --name pyrad python=3.8.13
conda activate pyrad
pip install -r environment/requirements.txt

# Install pyrad as a library
pip install -e .

# Install library with CUDA support. Change setup.py to `USE_CUDA = True` and then
python setup.py develop

# Running the test cases
pytest tests

Getting the data

Download the original NeRF dataset and put it in the following format.

- data/
    - blender/
        - fern/
        - lego/

Training a model

# Run with default config
python scripts/run_train.py

# Run with config changes
python scripts/run_train.py machine_config.num_gpus=1
python scripts/run_train.py data.dataset.downscale_factor=1

# Run with different datasets
python scripts/run_train.py data/dataset=blender_lego
python scripts/run_train.py data/dataset=friends_TBBT-big_living_room

# Run with different datasets and config changes
python scripts/run_train.py data/dataset=friends_TBBT-big_living_room graph.network.far_plane=14

Getting around the codebase

The entry point for training starts at scripts/run_train.py, which spawns instances of our Trainer() class (in nerf/trainer.py). The Trainer() is responsible for setting up the datasets and NeRF graph depending on the config specified. If you are planning on just using our codebase to build a new NeRF method or use an existing implementation, we've abstracted away the training routine in these two files and chances are you will not need to touch them.

The NeRF graph definitions can be found in nerf/graph/. Each implementation of NeRF is definined in its own file. For instance, nerf/graph/instant_ngp.py contains populates the NGPGraph() class with all of the appropriate fields, colliders, and misc. modules.

• Fields (nerf/fields/): composed of field modules (nerf/field_modules/) and represents the radiance field of the NeRF.
• Misc. Modules (nerf/misc_modules- TODO(maybe move to misc_modules? better organization)): any remaining module in the NeRF (e.g. renderers, samplers, losses, and metrics).

To implement any pre-existing NeRF that we have not yet implemented under nerf/graph/, create a new graph structure by using provided modules or any new module you define. Then create an associated config making sure __target__ points to your NeRF class (see here for more info on how to create the config). Then run training as described above.

Feature Documentation

1. Hydra config structure

2. Logging, debugging utilities

3. Benchmarking, other tooling

4. Running other repos with our data

# nerf-pytorch
cd external
python run_nerf.py --config configs/chair.txt --datadir /path/to/pyrad/data/blender/chair

# jaxnerf
cd external
conda activate jaxnerf
python -m jaxnerf.train --data_dir=/path/to/pyrad/data/blender/chair --train_dir=/path/to/pyrad/outputs/blender_chair_jaxnerf --config=/path/to/pyrad/external/jaxnerf/configs/demo --render_every 100

5. Speeding up the code

Documentation for running the code with CUDA. Please see https://github.com/NVlabs/tiny-cuda-nn for how to install tiny-cuda-nn.

pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch

To run instant-ngp with tcnn, you can do the following. This is with the fox dataset.

python scripts/run_train.py --config-name=instant_ngp_tcnn.yaml data/dataset=instant_ngp_fox

6. Setting up Jupyter

python -m jupyter lab build
bash environments/run_jupyter.sh