retraining
- noun
- /ˌriːˈtreɪnɪŋ/
- Specialized
:
reentrenamiento
- And then they're constantly in the background retraining that model, adding more information, refining the edges, the weights.
Like maybe you, like, maybe you can verify it for simple cases and then scale it up without retraining it somehow.
- Like maybe you, like, maybe you can verify it for simple cases and then scale it up without retraining it somehow.
Examples
-
Most of them were in the virtual lab retraining their models for areas they found were going bad on the physical track.
-
And during that retraining process that we got, we were basically told to tone it down a bit.
-
With ControlLogix, for example, it seems to be a retraining exercise every time a new release comes out.
Blog text (25) -
Using the SMART Gym, not only are we going to be able to use very specific retraining exercises, but we're going to give them markers.
-
And that's exactly what we're looking at is how we can use the retraining scheme as part of our response to the coronavirus.
-
So what we do through these games is we embed this technique for retraining your attention into the game so you can actually access it on a mobile device.
-
Although kappas remained above .75, a single retraining session was completed to address minimal drift issues.
Academic text (2007) -
The engineers implemented a new retraining strategy to improve the model's performance with the latest data.
-
To adapt to changing user preferences, the team scheduled regular retraining sessions for their predictive algorithms.
-
After noticing a decline in accuracy, the data scientists initiated a thorough retraining process on the existing machine learning model.
Synonyms
85
2.8
Data and process used to teach a computer program to find patterns
To teach or train someone again to refresh their skills
Antonyms
How Extensive
Surface Forms
Morphology
Etymology
Retraining comes from the parts re- and train, where re- means 'again' and train means 'to teach or practice'. In machine learning, retraining is teaching an existing model again with new data so it learns changed patterns and stays accurate.