Training images for gender and and age detection from face

Can someone please tell me if it is feasible to train my own set of facial images to detect gender and age without using any cloud architecture or paying some amount of money ?

And , on an average how many images do I need to train my model on ? This question arises since I have to take into account my computational facilities . I use a Windows OS ( Thinkpad) with 8 GB RAM .

Is it feasible for me to train my own model or I am better off to use a pretrained model ?