After completion of deeplearning.ai courses I am looking for personal project to further improve my understanding of AI and have some fun.<p>I find applications of convolutional networks and sequence models especially interesting.<p>One of the ideas I have is to build a command-line tool to classify my photos (mark photos with people, nature, cities, similar photos, photos with bad light / focus, ...).<p>Could you suggest progression through such project with high-level tasks ordered by difficulty?<p>Other project suggestions are also welcome!
Most straightforward progression:<p>1. establish desired categories<p>2. group existing photos into these categories<p>3. finetune a pretrained model on these categories<p>You can also partially automate step 2 using a pretrained model, but in this case some of your desired classes should be similar to (some of) the classes that model was trained on.