Coursera course requires version 1.8.3, which is wrong

graphlab-create 1.8.3.txt

(data2) C:\Users\hughdbrown>pip install graphlab-create==1.8
Collecting graphlab-create==1.8
  Could not find a version that satisfies the requirement graphlab-create==1.8 (from versions: 2.1)
No matching distribution found for graphlab-create==1.8

(data2) C:\Users\hughdbrown>pip install graphlab-create==1.8.3
Collecting graphlab-create==1.8.3
  Could not find a version that satisfies the requirement graphlab-create==1.8.3 (from versions: 2.1)
No matching distribution found for graphlab-create==1.8.3


The only hit that comes up for graphlab-create on pypi is version 2.1:

https://pypi.python.org/pypi?%3Aaction=search&term=graphlab-create&submit=search

pypi page for graphlab-create mentions no previous versions for 2.1:
https://pypi.python.org/pypi/GraphLab-Create/2.1

## Stackoverflow questions
http://stackoverflow.com/questions/39042052/not-able-to-get-version-1-8-3-of-graphlab-create-for-my-mooc-course

## Other pypi mirrors
https://gemnasium.com/pypi/GraphLab-Create

# Other questions

https://www.coursera.org/learn/ml-classification/discussions/all?q=1.8.3

https://www.coursera.org/learn/ml-classification/discussions/all/threads/av_eoGhzEeaMKQ4sYRHqqQ

> Version 1.8.3 is no longer available, so it is not possible to downgrade to that particular version.
> 
> Those questions asking you to install version 1.8.3 were meant to ensure that students did not use version 1.8.4, which produced different numbers than expected and also had a performance regression, if I recall correctly.
> 
> Any version of GraphLab Create other than 1.8.4 should work. There is an open ticket on GitHub to remove the instructions to install 1.8.3: https://github.com/learnml/machine-learning-specialization/issues/17

> I just upgraded GraphLab Create on my computer to the latest version, 2.1, and re-ran all cells of the notebook. I get the same sets of top 20 most positive and most negative reviews.
> 
> One thing to check is that you are using the topk() method. If you sort() and take the first k, this is known to produce different results than calling topk().