@fly4dat 737,570 robotaxis
-
-
-
Correction: 0 robotaxis with this hardware, but a few with Waymo systems.
- Još 8 drugih odgovora
Novi razgovor -
-
-
5 pillars of Tesla fleet learning: 1. Active learning for computer vision (CV) 2. Weakly supervised learning for CV 3. Self-supervised learning for CV 4. Self-supervised learning for prediction 5. Imitation learning (and maybe RL) for planninghttps://link.medium.com/WLNQGLCC02
-
So last month. Have u heard of the self learning breakthroughs and unlabelled image training with the Noisy Student model (Q. Le, et al)? See Nov. paper on http://Arxiv.org where all AI papers are published.
- Još 5 drugih odgovora
Novi razgovor -
-
-
Thanks for updating so soon with Q4 numbers. It's amazing how few people in the analyst world seem to understand how IMPORTANT Tesla's data lead is. RIP other autos.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
I wish Tesla would tell us the best way to help them teach the machine.
-
The best way is to park the car somewhere with WiFi overnight. If you don’t do this, Tesla cannot use your car’s data for fleet learning.
- Još 5 drugih odgovora
Novi razgovor -
-
-
#Tesla killing it with the Data and ML -
Perhaps that's not the right choice of word. ;-)
Kraj razgovora
Novi razgovor -
-
-
it's a multilayered question. Tesla the car does not 100% But it can generate a report in some cases. The report then may get transmitted to Tesla. And if somebody at Tesla (or some process even) decided to take a look, finds it interesting enough and works on it, it may get inco
- Još 1 odgovor
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.