Interesting finding but the 10x is exaggeration here. ShuffleNet V2 with 597M FLOPs gets 75.4% top-1 accuracy on image net and the EfficientNet with 390M FLOPs gets 76.3% top-1. The 10x is not compared with the state-of-the-art efficient models.
EfficientNet is an open source library that uses a new compound model scaling method and leverages recent progress in #AutoML to improve #neuralnetwork scaling techniques, achieving state-of-the-art accuracy with up to 10x better efficiency. Learn more ↓https://goo.gle/2VXWtDx
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
If you are still using ResNet or Inception, please consider giving it a try! Open source link:https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/ …
- Još 1 odgovor
Novi razgovor -
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
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.