Mastering Time and Text: RNNs, LSTMs, and the Magic of Word Embeddings

Up until this point, we have operated in a static world. Whether we were predicting housing prices or running convolutions over an image of a dog, the inputs were fixed in time and space. But what if