Rahul Bhargava
Rahul Bhargava
In our testing the current code produces unreliable results when tested on Wikipedia articles. Sometimes it returns a data, sometimes it doesn't. Wikipedia articles are constantly updated, so @coreydockser and...
I'm trying to understand some poor precision results we are getting when using the QuoteAnnotator on a news corpus. I'm hoping to understand and tune the sieves to improve precision,...
### Description of problem Deploying from a local tag to master on remote host reported a super odd error: ``` -----> WARNING: deploy did not complete, you must push to...
We found that the `top_image` for a noticeable number of stories in a sample of news we were working on returned favicons. This happened on stories from popular and large...
I haven't done rigorous testing yet, but has anyone else noticed v2.0.0 being about half the speed of v1.1? I updated our [CLIFF tool to use it](https://github.com/c4fcm/CLIFF/releases/tag/v2.0.0), but it is...
NYT offers an API endpoint that suggests matching entities from their list of ~15k people and ~7k orgs. I wonder if we could use this to build a simple substitution-based...
The commercial https://newsapi.org site appears to have ~30k sources categorized by country of publication and language. Might be worth scraping from their `sources` endpoint and importing that metadata for any...
Rewrite the API from the ground up. Some goals: * python instead of perl * use an api generator to help? * better error reporting * more restful (placeholder issue...
A collaborator reports that doing a story list with `sort=mc.SORT_PUBLISH_DATE_DESC` doesn't work. Sort by `random` and `processed_stories_id` do work fine. Can you check on if this works and is still...
It is a little hard to figure out exactly what each "app" container is for, even thought the names are descriptive. I think it'd be helpful to have a high-level,...