Hello, everyone!
I just finished this project and want to know if I mistaken anything. Feel free to give me some feedback. Here’s the link to my project Google Colab
Thank you 
Hi @IndraSubagja
Honestly I’m a lot more familiar with ML in Scikitt-Learn. But as far as I can see your code looks good for me. (Based also on the instructions). You made the correct steps,
Feature Engineering, Split the data (validation_split=0.2), Training an Testing Accuracy.
So its seems ok.
Tip: Note that the Learning Rate is not fixed by the instructions, so if in the future you want to improve your model, you can made use of this detail (A tool called Learning Rate Finder)
Congratulations for your project.
@AndyG Thanks! Honestly, I’m not sure with what I’m doing, especially when create model. I can’t decide the loss function, optimizer, and also the way I set the layers and each parameters
Well, not a big deal, you just need a bit of Machine Learning Theory. Also, choosing the hyperparameters will depend of your criteria (what do you want to obtain from the model) and the model specificity.
As I said, I’m not a Keras expert, but taking the exercise intructions as a base, the coding of the model seems ok.
If you want to learn a bit more, I recommend to check this crash course:
Thanks, I’ll check it out
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.