https://linktr.ee/rikimatsumoto
wow thanks google Gemini for predicting the future 😐
(I am gonna pass out if this is actually spot on)
wow thanks google Gemini for predicting the future 😐
(I am gonna pass out if this is actually spot on)
I.e., if the model has a good (low) RMSE but predicts a noticeably different distribution, it might be overfitting the center and neglecting the tails. Distributional metrics help identify this.
I.e., if the model has a good (low) RMSE but predicts a noticeably different distribution, it might be overfitting the center and neglecting the tails. Distributional metrics help identify this.
www.brookings.edu/articles/tra...
Massive yoy percent increases in duties for March 2025 among top trading partners.
www.brookings.edu/articles/tra...
Massive yoy percent increases in duties for March 2025 among top trading partners.
Advanced release Census data shows approx 50% yoy increase in U.S. consumer goods imports in March.
Advanced release Census data shows approx 50% yoy increase in U.S. consumer goods imports in March.
- largest increases in housing, durables, energy, and food
- expenditure on clothing hasn’t recovered since 2020
- culture/recreation almost back up to 2019 levels
- largest increases in housing, durables, energy, and food
- expenditure on clothing hasn’t recovered since 2020
- culture/recreation almost back up to 2019 levels
Arguably, it was 80%:
1) mitigating the impact of AI driven energy consumption, and;
2) AI for mitigating climate risks (better seed varieties)
Sparse on solving climate change or hitting emissions goals, overall bearish on that.
Arguably, it was 80%:
1) mitigating the impact of AI driven energy consumption, and;
2) AI for mitigating climate risks (better seed varieties)
Sparse on solving climate change or hitting emissions goals, overall bearish on that.
I asked a general Llama3 to evaluate and grade its own output. As the K top matches and similarity score increases, quality of the output drops precipitously!
I asked a general Llama3 to evaluate and grade its own output. As the K top matches and similarity score increases, quality of the output drops precipitously!