search

QuantEcon / lecture-python-programming

Source files for https://python-programming.quantecon.org
BSD 3-Clause "New" or "Revised" License
18 stars 13 forks source link

Add Dask discussion and GPU discussion to parallelization lecture #19

Open jstac opened 4 years ago

jstac commented 4 years ago

A new parallelization lecture was added in QuantEcon/lecture-source-py#719

At the bottom are commented out headings for Dask and GPUs.

These can be added sequentially while the lecture is live, probabily starting with Dask.

@mmcky, I'll leave this up to you, as discussed. The lecture is already listed as co-authored with you. It might be good to read the whole lecture and the ones mentioned within to get a sense of the flow of discussion and know what has already been said.

(The GPU discussion can be put aside if you like but it only needs to be a few lines at this stage, with no example code for obvious reasons.)

mmcky commented 4 years ago

thanks @jstac I will have a look at this on Monday.

jstac commented 4 years ago

This might be useful to you @mmcky: https://xarray.pydata.org/en/stable/dask.html

jstac commented 4 years ago

And Dask-controlled GPUs!

https://devblogs.nvidia.com/accelerating-python-for-exotic-option-pricing/

Such a fun project.

I think we should actually have a separate lecture on GPUs with a warning that it won't run on regular machines. Or a mini-series of lectures specifically on using GPUs? Who's in?