Marc Chevrette

Results 7 comments of Marc Chevrette

Overall discussion for Tip 2

Could add a line about this reference https://academic.oup.com/nar/article/47/18/e110/5545735 here; it makes the point to benchmark their new DL model against non-machine-learning methods. The goal is to identify biosynthetic gene clusters...

Fun example that highlights over fitting (not DL specific)

Was thinking something similarly obvious could be a panel B for our current figure. Just a thought.

Make sure your problem is a good fit for deep learning.

To this end: https://twitter.com/FinlayM/status/1084150174300164096?s=19

Fairly validate your model with proper test/train data

Could be merged as is proposed elsewhere but I feel that this is important enough to be its own rule

Fairly validate your model with proper test/train data

Completely agree will all. How might we rephrase this to make it more DL-specific? Or rather should we leaveit more ML in general and add DL-specific caveats as we explain?...

There's more than one way to do it: find the learning that's right for you (and your data)

Could possibly be linked to a figure where data (or multiple examples) through different classifiers (and different parameters) gives different precision/recall. I think a visual like this might be very...

There's more than one way to do it: find the learning that's right for you (and your data)

Similarly, some text on the importance of parameters into your models (eg alpha, number of layers, etc etc), what they mean, and how they can influence how you might interpret...