Best choice for my use case?

[Mitch0S]

G'day, how's it going?

I've just started looking into machine learning stuff, and stumbled upon this, looks awesome!

I just want to know what kind of methods I should use for the following:

• Text identification (Spam checker for example)
• Image analysis (Detects whether the image given after training is a male or female)

Kind regards,

Machine-Learning newbie, Mitch!

[ddbourgin]

Hi Mitch! Sorry for the slow response. Yours is a big question, and there's no single right answer unfortunately (I hate this kind of "response," but it's true). A few pointers:

• Text identification: The simplest version of this problem uses models from supervised classification. This could run the gamut from simple linear classifiers (e.g., logistic regression, naive Bayes) to deep neural networks (e.g., recent transformer models). For spam classification in particular, the canonical (but certainly not the most performant) example is a naive Bayes classifier.

• Image analysis: Again, huge swath of models, although most meaningful models these days are variations on convolutional neural networks (of which the Conv2d layer forms the backbone). For classification, you would generally be relying on some form of cross-entropy loss. Unfortunately documentation on all of this is not as developed as I'd like it to be, but I think there's a pretty good correspondence between the operations here and those implemented in other popular deep learning frameworks (e.g, Keras/Tensorflow, PyTorch), so you might check there as well as you get familiar :)