IFT 3710/6759 - Project development

name: title
class: title, middle

## IFT 3710/6759
## Projets (avancés) en apprentissage automatique

### .gray224[March 9th, 2022 - Session 16]
### .gray224[Project development]

.center[
<a href="http://www.umontreal.ca/"><img src="../../../assets/images/slides/logos/udem-white.png" alt="Mila" style="height: 6em"></a>
]
<br><br>
Álex Hernández-García (he/il/él)

.footer[[alexhernandezgarcia.github.io](https://alexhernandezgarcia.github.io/) | [alejandro.hernandez.garcia@umontreal.ca](mailto:alex.hernandez-garcia@mila.quebec) | [@alexhdezgcia](https://twitter.com/alexhdezgcia)] [![:scale 1em](/assets/images/slides/misc/twitter.png)](https://twitter.com/alexhdezgcia)

---

## Format of the class and objective

The first half an hour of the class will review some guidelines about the project development and the use of the cluster. The rest of the class will b e for project discussions.

Outline:

* Good practices for receiving feedback on your project
* Data sets in the cluster
* General Q&A
* 17:00 - 17:30: Downscaling Global Climate Models
* 17:30 - 18:00: Photovoltaic power and solar radiation forecasting (x2)
* 18:00 - 18:30: WILDS: Distribution shifts in the wild

---

## Good practices for receiving feedback on your project

Make it easier for your mentor and you will get the most of out them:

* Prepare a small slide deck for each meeting:
	* Progress
	* Who made what
	* Figures
	* Challenges
* Communication: agree with your mentor on the best communication channel.
	* A private Discord channel can be created for your team and your mentor.
* Discord channels can be created for specific topics
* Email communication:
    * Add on Cc all the members of the team.
    * Add an identifier of the team to the subject of the email.

---

## Data sets in the cluster

* Storage space in the cluster is limited.
* Having multiple copies of the data sets is both inefficient and a potential source of inconsistencies.
* Some locations in the cluster are optimised for data transfer.

Data sets used in the projects should be stored in the dedicated directory for data sets:

* `/project/def-sponsor00/datasets`
* `~/projects/def-sponsor00/datasets`

The two paths above point to the same disk location.

Please be **extremely careful** when dealing with this directory since it is shared by all of you!

---

## Data sets in the cluster
### What to do with the data

1. Copy your data set into `/project/def-sponsor00/datasets`
2. Let me know
3. Take extreme care with the data in this directory

---

name: title
class: title, middle

## IFT 3710/6759
## Projets (avancés) en apprentissage automatique

### .gray224[March 9th, 2022 - Session 16]
### .gray224[Project development]

.bigger[.bigger[.highlight1[Questions, doubts, concerns, comments?]]]

.center[
<a href="http://www.umontreal.ca/"><img src="../../../assets/images/slides/logos/udem-white.png" alt="Mila" style="height: 6em"></a>
]

Alex Hernández-García (he/il/él)