przemb

"""
In this script, we demonstrate how to use DINOBot to do one-shot imitation learning.
You first need to install the following repo and its requirements: https://github.com/ShirAmir/dino-vit-features.
You can then run this file inside that repo.

There are a few setup-dependent functions you need to implement, like getting an RGBD observation from the camera
or moving the robot, that you will find on top of this file.
"""
import torch
import numpy as np 

przemb

Reproducibility

ML experiments may be very hard to reproduce. You have a lot of hyperparameters, different dataset splits, different ways to preprocess your data, bugs, etc. Ideally, you should log data split (already preprocessed), all hyperparameters (including learning rate scheduling), the initial state of your model and optimizer, random seeds used for initialization, dataset shuffling and all of your code. Your GPU is also should be in deterministic mode (which is not the default mode). For every single model run. This is a very hard task. Different random seed can significantly change your metrics and even GPU-induced randomness can be important. We're not solving all of these problems, but we need to address at least what we can handle.

For every result you report in the paper you need (at least) to:

1. Track your model and optimizer hyperparameters (including learning rate schedule)
2. Save final model parameters
3. Report all of the parameters in the pap