Understanding Seq2Seq Neural Networks – Part 7: Generating the Output with Softmax

In the previous article, we were transforming the outputs to the fully connected layer. A fully connected layer is just another name for a basic neural network. This fully connected layer has two inputs, corresponding to the two values that come from the top layer of LSTM cells. It has four outputs, one for each token in the Spanish vocabulary. In between, we have connections between each input and output, each with their own weights and biases. Then, we run the output of the fully connected layer through a Softmax function to select the output word. Now, going back to the full Encoder–Decoder model: We can see that the output from the Softmax function is “Vamos,” which is the Spanish translation for “Let’s go.” We will continue further in the next article. Looking for an easier way to install tools, libraries, or entire repositories? Try Installerpedia: a community-driven, structured installation platform that lets you install almost anything with minimal hassle and clear, reliable gu