This code requires at least Python 3.9 and [PyTorch 1.11](https://pytorch.org/get-started/locally/). Additional dependencies may be installed via `pip -r requirements.txt`.
This code requires at least Python 3.9 and [PyTorch 1.11](https://pytorch.org/get-started/locally/).
Note that Tensorflow is required to load SRT's MultiShapeNet data, though the CPU version suffices.
Additional dependencies may be installed via `pip -r requirements.txt`. Note that Tensorflow is
required to load OSRT's MultiShapeNet data, though the CPU version suffices.
Rendering videos additionally depends on `ffmpeg>=4.3` being available in your `$PATH`.
Rendering videos additionally depends on `ffmpeg>=4.3` being available in your `$PATH`.
...
@@ -37,17 +38,26 @@ To train on multiple GPUs on a single machine, launch multiple processes via [To
...
@@ -37,17 +38,26 @@ To train on multiple GPUs on a single machine, launch multiple processes via [To
Checkpoints are automatically stored in and (if available) loaded from the run directory. Visualizations and evaluations are produced periodically.
Checkpoints are automatically stored in and (if available) loaded from the run directory.
Check the args of `train.py` for additional options. Importantly, to log training progress, use the `--wandb` flag to enable [Weights & Biases](https://wandb.ai).
Visualizations and evaluations are produced periodically. Check the args of `train.py` for
additional options. Importantly, to log training progress, use the `--wandb` flag to enable [Weights