Continue to Train an Already Trained Keras Model with New Data

Share this video with your friends

Send Tweet

As we get new data, we will want to re-train our old models with that new data. We’ll look at how to load the existing model, and train it with new data, and then save the newly trained model.

Vatsaev Aslan
Vatsaev Aslan
~ 6 years ago

Could've talked about transfer learning a little bit

Chris Achard
Chris Achard(instructor)
~ 6 years ago

Vatsaev: this was meant to be a more "intro" level course, but Yes! transfer learning would be a great topic in a follow on course. Using an already trained model is definitely where a lot of the power of using neural networks comes into play.

Shawn Wang
Shawn Wang
~ 6 years ago

seconding transfer learning

Eliezer Steinbock
Eliezer Steinbock
~ 6 years ago

Would love to see a follow up to this course! I want more!

Rene
Rene
~ 6 years ago

yes, a follow up would be nice. how to learn with different counts of inputs ? e.g. short and long text ?

Alplob
Alplob
~ 5 years ago

Thanks a lot for that very practical course, I am big fan of it! If there is a new course about Keras, I would be very interested about handling multiple time series data like those about Antarctica weather shown in the course 5.