Yes, that's the traditional way. I drew it differently on purpose, because some applications of machine learning, like training a logistic regression on the Iris flower dataset, is hardly AI in my opinion.
It could be, but then it would be a bit too specific to a given model. As a general performance measure that is also easy to interpret, I'd go with the percentage of digits classified correctly
Starting along this your book python machine learning......
Hope to get a crystal clarity on concepts and logic behind them.
Great content! Thank you Prof. Raschka
Glad to hear you are liking it!
Thank you for sharing your lectures with us!
You are welcome, I hope they turned out to be useful to you!
Wow - this quote from 1959 is so up-to-date. I'm amazed.
“Field of study that gives computers the ability to learn without being explicitly programmed.” -Arthur Samuel.
Agreed! It's really fascinating to think that this was 63 years ago!
wait, shouldn't that second diagram be the reverse of the first? With AI a subset of ML rather than partly outside it?
Yes, that's the traditional way. I drew it differently on purpose, because some applications of machine learning, like training a logistic regression on the Iris flower dataset, is hardly AI in my opinion.
Hi Sabastian, is it loss functions are performance measure of P?
It could be, but then it would be a bit too specific to a given model. As a general performance measure that is also easy to interpret, I'd go with the percentage of digits classified correctly
Thanks...😊
I bought your book. But I prefere your videos is it bad??