fastml extra

@fastml_extra

Everything related to machine learning and data science in general. Follow for notifications about new posts and for data visualizations.

ಜೂನ್ 2013 ಸಮಯದಲ್ಲಿ ಸೇರಿದ್ದಾರೆ

@fastml_extra ತಡೆಹಿಡಿಯಲಾಗಿದೆ

ನೀವು ಖಚಿತವಾಗಿಯೂ ಈ ಟ್ವೀಟ್‌ಗಳನ್ನು ನೋಡಲು ಬಯಸುವಿರಾ? ಟ್ವೀಟ್‌ಗಳನ್ನು ನೋಡುವುದು @fastml_extra ಅವರನ್ನು ತಡೆತೆರವುಗೊಳಿಸುವುದಿಲ್ಲ.

  1. Retweeted
    ಫೆಬ್ರ 25

    A mural is purposely painted upside down to reflect right side up in the water

  2. Retweeted
    ಫೆಬ್ರ 25

    New release 1.0.0. Switch to semantic versioning. Major new features. Breaking change in .

  3. ಫೆಬ್ರ 25

    Man, stop yapping and proceed with getting rich with those stock market predictions of yours.

  4. Retweeted
    ಫೆಬ್ರ 23

    Prophet is an open-source forecasting package implemented in both Python and R, used in production at Facebook:

  5. Retweeted
    ಫೆಬ್ರ 23

    Shake-Shake regularization code claims 2.72% on CIFAR-10. Fun - add more stochastic, even "break" backprop.

  6. ಫೆಬ್ರ 24

    "Why would I use it?" That's a very good question. Well, if you only had logistic regression and a decision tree as an alternative...

  7. Retweeted
    ಫೆಬ್ರ 24

    New music video from featuring around 3:14 or so. Pretty cool.

  8. ಫೆಬ್ರ 24

    What would you do with unlimited computing power? "Two classifiers at once" and more

  9. ಫೆಬ್ರ 24
  10. ಫೆಬ್ರ 24
  11. Retweeted
    ಫೆಬ್ರ 21

    Has anyone tried this? 'Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks'

  12. Retweeted
    ಫೆಬ್ರ 21

    Wages of Wins: Could an Amateur Make Money from Match Outcome Predictions? Use Naive Bayes and wait for mid-season!

  13. Retweeted
    ಫೆಬ್ರ 21

    Char2Wav: Great progress towards Speech Synthesis without requiring years of language-specific feature engineering!

  14. Retweeted
    ಫೆಬ್ರ 21

    Wow. You can draw a picture and this website will turn it into a cat!

  15. Retweeted
    ಫೆಬ್ರ 20

    Ooh no IBM Watson having another fail and loosing an important client. 😒

  16. Retweeted
    ಡಿಸೆಂ 12,2016

    Every time I get nervous about AI, I picture my Roomba stuck under this table.

  17. Retweeted
    ಫೆಬ್ರ 21

    I like this paper a lot. Refreshing to see alternative methods to policy search, away from typical resource-heavy and time-consuming DeepRL.

  18. Retweeted
    ಫೆಬ್ರ 18

    Audio Super-Resolution using Neural Networks I was wondering how long this would take to happen! No samples yet.

  19. Retweeted
    ಫೆಬ್ರ 18

    Generating plausible colour palettes via GANs ()

  20. Retweeted
    ಫೆಬ್ರ 17

    Contrary to popular belief, Deep Learning is theoretically grounded. We just don't know what the theory is yet.

ಲೋಡಿಂಗ್ ಸಮಯ ಸ್ವಲ್ಪ ತೆಗೆದುಕೊಳ್ಳುತ್ತಿರುವಂತೆನಿಸುತ್ತದೆ.

Twitter ಸಾಮರ್ಥ್ಯ ಮೀರಿರಬಹುದು ಅಥವಾ ಕ್ಷಣಿಕವಾದ ತೊಂದರೆಯನ್ನು ಅನುಭವಿಸುತ್ತಿರಬಹುದು. ಮತ್ತೆ ಪ್ರಯತ್ನಿಸಿ ಅಥವಾ ಹೆಚ್ಚಿನ ಮಾಹಿತಿಗೆ Twitter ಸ್ಥಿತಿಗೆ ಭೇಟಿ ನೀಡಿ.

    ಇದನ್ನೂ ಸಹ ನೀವು ಇಷ್ಟಪಡಬಹುದು

    ·