Amsterdam Internships - Example README

Explain in short what this repository is. Mind the target audience. No need to go into too much technical details if you expect some people would just use it as end-users and don't care about the internals (so focus on what the code really does), not how. The How it works section below would contain more technical details for curious people.

If applicable, you can also show an example of the final output.

---

Project Folder Structure

Explain briefly what's where so people can find their way around. For example:

There are the following folders in the structure:

1. resources: Random nice resources, e.g. useful links
2. src: Folder for all source files specific to this project
3. scripts: Folder with example scripts for performing different tasks (could serve as usage documentation)
4. tests Test example
5. media: Folder containing media files (icons, video)
6. ...

OR

Or use something like tree to include the overall structure with preferred level of detail (-L 2 or -d or -a...)

├── media --> you can still add comments and descriptions in this tree
│   └── examples
├── resources --> a lot of useful links here
├── scripts
├── src --
└── tests

If you are lacking ideas on how to structure your code at the first place, take a look at CookieCutter

---

Installation

Explain how to set up everything. Let people know if there are weird dependencies - if so feel free to add links to guides and tutorials.

A person should be able to clone this repo, follow your instructions blindly, and still end up with something fully working!

1. Clone this repository:

   git clone https://github.com/Amsterdam-Internships/InternshipAmsterdamGeneral

2. If you are using submodules don't forget to include --recurse-submodules to the step above or mention that people can still do it afterwards:

   git submodule update --init --recursive

3. Install all dependencies:

   pip install -r requirements.txt

---

Usage

Explain example usage, possible arguments, etc. E.g.:

To train...

$ python train.py --some-importang-argument

If there are too many command line arguments, you can add a nice table with explanation (thanks, Diana Epureano!)

Argument	Type or Action	Description	Default
--batch_size	int	Batch size.	32
--device	str	Training device, cpu or cuda:0.	cpu
--early-stopping	store_true	Early stopping for training of sparse transformer.	True
--epochs	int	Number of epochs.	21
--input_size	int	Input size for model, i.e. the concatenation length of te, se and target.	99
--loss	str	Type of loss to be used during training. Options: RMSE, MAE.	RMSE
--lr	float	Learning rate.	1e-3
--train_ratio	float	Percentage of the training set.	0.7
...	...	...	...

Alternatively, as a way of documenting the intended usage, you could add a scripts folder with a number of scripts for setting up the environment, performing training in different modes or different tasks, evaluation, etc (thanks, Tom Lotze!)

---

How it works

You can explain roughly how the code works, what the main components are, how certain crucial steps are performed...

---

Acknowledgements

Don't forget to acknowledge any work by others that you have used for your project. Add links and check whether the authors have explicitly stated citation preference for using the DOI or citing a paper or so. For example:

Our code uses YOLOv5