freedomofpress / fingerprint-securedrop

A machine learning data analysis pipeline for analyzing website fingerprinting attacks and defenses.
GNU Affero General Public License v3.0
29 stars 9 forks source link

Add data SITREP #103

Closed redshiftzero closed 7 years ago

redshiftzero commented 7 years ago

Here's an ipython notebook running some quick numbers on our data thus far: TL;DR we have 200k examples to 5.1k unique onion services. w00t

No reviewing requested here - this is a "check it out if you are interested PR"

Also one can steal this notebook and re-run this at a later date if you want to regenerate these numbers and plots to see how things are going 😉, you just need $PGPASS defined

conorsch commented 7 years ago

Evidently I could use a bit of a refresher on how to view this notebook locally. Is jupyter notebook notebooks/data_crawling_status.ipynb sufficient?

redshiftzero commented 7 years ago

Yep that will work. Conveniently Github will also render the notebook if you navigate to it: https://github.com/freedomofpress/fingerprint-securedrop/blob/add-data-state-notebook/notebooks/data_crawling_status.ipynb

conorsch commented 7 years ago

Excellent report, thank you for sharing this! It's a shame we had such low collection rates for a while, but glad to hear the bits are flowing now.

coveralls commented 7 years ago

Coverage Status

Coverage remained the same at 72.727% when pulling 9de8d9a00f0dfce0eeba5ff0eb044c524ea6f523 on add-data-state-notebook into 33be2d7d550a222f9d4870f25dbaf4fda4e86b7c on master.

psivesely commented 7 years ago

Will review tomorrow!

redshiftzero commented 7 years ago

Feel free to peruse the notebook at your leisure 😃, but a review is not necessary here (see PR description), so I'm gonna go ahead and merge this one in. However I am about to submit another PR in this repo for you to review 😉