×
ECE594
Home
Schedule
Piazza
Gradescope
Logistics
Projects
Resources
 

ECE 594: Mathematical Models of Language

Resources

1. Campus Cluster for computing needs

Our course will use the shared Engineering Instructional partition on the Campus Cluster. Here are some details about the eng-instruction partition:

eng-instruction - consisting of:
+ 8 dual-socket Intel Xeon E5-2680v4 Broadwell CPU nodes w/ 64GB RAM (224 CPU total cores) w/EDR InfiniBand interconnects 
+ 2 dual-socket Intel Xeon E5-2680v4 Broadwell GPU nodes w/ 256GB RAM & 1 NVIDIA P100 GPU (56 total GPU cores) w/EDR InfiniBand interconnects
+ 1 dual-socket Intel Xeon E5-2680v4 Broadwell GPU node w/ 256GB RAM & 4 NVIDIA P100 GPU (224 total GPU cores) Ethernet-only interconnect

Please refer to https://campuscluster.illinois.edu/resources/docs/start/
for information on how to get started. 
https://campuscluster.illinois.edu/resources/docs/user-guide/ contains more comprehensive usage information including storage policies and locations (https://campuscluster.illinois.edu/resources/docs/user-guide/#fs)

If your jobs require the use of GPU’s you can request the available GPU resources by using the feature flags as outlined here:

https://campuscluster.illinois.edu/resources/docs/user-guide/#gpus

The College of Engineering has some shared storage available; however it is limited, so we ask that you be considerate in its use. Please contact the Technical Representatives at techrep@engr.illinois.edu in order to be granted access to engineering storage for your course.

If you require large amounts of storage (>1 TB) we ask that you consult with the Technical Representatives to discuss other solutions.

Because you have been added as a user of the Campus Cluster, you should be automatically enrolled in the ICCP users mailing list, where you'll receive important updates about outages or preventative maintenance (PM) down times related to the campus cluster.

Any technical issues you may experience with the cluster should be addressed to the cluster administrators at: help@campuscluster.illinois.edu

 

2. Project-related resources

  • A good overview of the state of NLP is available at NLP progress, thanks to the extremely laudable efforts of Sebastian Ruder.

  • If you are looking for models with implementations to experiment with, check out the papers with code. 

  • All *NLP/*CL papers from the major NLP conferences are available here.

  • For quick and easy hacky NLP solutions, try this site.

  • For inspirations, check out what students from other institutions have been doing!