When it comes to language learning, oftentimes "improvement" and "the feeling of improvement" don't correlate very well, and this gets many learners into trouble.
-
-
It's better to continually consume new input, in order to provide your brain with as much i+1 input as possible. If you take some random passage and study it to death, most of it won't actually be i+1, so you won't actually be able to acquire it.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
The same problem is also seen in machine learning and called overfitting. If training dataset is too small, the model does very well in training set performance but when tested on real samples it fails badly. The best way to fix it is by having more and varied data like you say.
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.