prashp

8 ms·

1. 61.▲
   Blender Graphics: Lenses and Optics(arachnoid.com)

   11 points by prashp 4y ago|4 comments
2. 62.▲
   by prashp 4y ago

   Is Gwern [1] actually Stephen Wolfram's alter ego? Two sides of the same coin [1] https://www.gwern.net/
3. 63.▲
   by prashp 4y ago

   Imagine sacrificing your health and time with family for Twitter of all things, and then doubling down on it publicly.
4. 64.▲
   by prashp 4y ago

   You like your lobsters
5. 65.▲
   by prashp 4y ago

   This is academic peer-review in a nutshell.
6. 66.▲
   by prashp 5y ago

   If you're not white, don't move there. The Swiss culture is very unwelcoming to newcomers.
7. 67.▲
   by prashp 5y ago

   I love Avatar, but the analogies in that notebook were quite difficult to follow!
8. 68.▲
   by prashp 6y ago

   This seems like a thinly veiled advertisement for Gradio. This is an otherwise trivial observation of convolutional neural networks, given the rapid pace of the field in general
9. 69.▲
   by prashp 6y ago

   How does debugging work in Deepnote vs Jupyter?
10. 70.▲
    by prashp 6y ago

    What about providing some empirical evidence on how to choose a network? It's not enough to list a few alternative architectures - how are readers supposed to know which ones are worth trying first? This seems to be a problem in deep l
11. 71.▲
    by prashp 7y ago

    How is accuracy calculated?
12. 72.▲
    by prashp 7y ago

    Cool! I use my old laptop as a cat camera too. I use the integrated webcam to take a photo every 5 seconds, and then I send it through a lightweight CNN to classify of there is a cat in the photo. If there is I get an email with the photo a
13. 73.▲
    by prashp 7y ago

    I disagree with you. Weak baselines are inappropriate, especially when claiming 'state-of-the-art' results, which is very common in ML literature these days. If weak baselines are used as proof of superiority for a particular meth
14. 74.▲
    by prashp 8y ago

    TuriCreate also provides good tutorials on doing this: https://apple.github.io/turicreate/docs/userguide/object_det...
15. 75.▲
    by prashp 8y ago

    Actually, this is object detection because there is a bounding box which is generated localizing the object(s). Classification would be if each frame returned a class and class score without the bounding box.