Fri, Mar 16
Thu, Mar 15
The recommended order for review should be - 18, 20, 19
Final resolution done by using a wrapper function - https://github.com/wiki-ai/revscoring/pull/394
Tue, Mar 13
@Ragesoss there's ongoing work around topic modeling for English Wikipedia using WikiProject topics as bases. If Education Program Dashboard has some similar categorization of articles around pre-defined topics, a similar model can be built to predict topics as well as recommend them. Let me know if you wanna talk more about it.
Refer to the gist in the first comment for the code changes that make it multiprocessing friendly.
Sat, Mar 10
Test code for benchmarking using an word2vec as an external module contained in english_vectors:
from multiprocessing import Pool, cpu_count import functools from revscoring.dependencies import solve from revscoring.datasources.meta import vectorizers from revscoring.features.meta import aggregators from revscoring.languages import english from revscoring.languages.english_vectors import google_news_kvs from revscoring.datasources import revision_oriented
Test code for benchmarking vectorizers with a global keyed_vector in the file:
Thu, Mar 8
with wordvectors blockers now cleared, building drafttopic model on ores-stat-01
Tue, Feb 27
Feb 13 2018
Feb 5 2018
Jan 29 2018
Jan 22 2018
A common use case of fetch_text is augmenting the dataset with X info from Y api. This will address:
Jan 17 2018
The binary *was* on ores-misc-01 which is now nuked. I'll upload it to ores-staging-01 from my system again from where it can be put somewhere public.
Jan 16 2018
I've taken backup of the tuning reports, and the GradientBoosting and RandomForest models.
Dec 22 2017
Dec 20 2017
Dec 11 2017
Nov 29 2017
Nov 28 2017
We now have a dataset at figshare - https://doi.org/10.6084/m9.figshare.5640526.v1 \o/
Nov 22 2017
Nov 21 2017
Nov 6 2017
Nov 4 2017
Nov 3 2017
Could free up 2.2G more...
Removed 800MB of my stuff which included cached models and datasets.
Oct 30 2017
Oct 16 2017
Published on ORES account of figshare.