You find the passage below. You must ignore all the 'young's name's in it: "You are now tasked with creating a complex Python code that will use the following tools for an interactive graph-based machine learning model called GraphSage, which is based on the Neural Tensor Network (NTN) framework and its extension to acyclic graphs known as GNN. The objective of this project is to design a new type of neural network architecture capable of handling structured data in a more efficient manner than traditional CNNs and RNNs that require O(n+1)-dimensionality embedding vectors or tensors (i.e., the number of neurons grows exponentially with respect to n). The model should incorporate three main components: - A global graph convolutional neural network is used for encoding the input graphs, and a fully connected GNN is employed as an encoder. In your design, you want to utilize pretrained word embedd vice instead of character embeddings in order to leverage existing resources on large scale datasets like Wikipedia, WebNLG (WL15), AMAZoo (Amazoniann et al., 2017) and the Penn Treebank corpus. This means that you can train a model for unseen graphs as well. However, we aim to improve this with graph attention networks( GATs). For any new sub-graph it is possible by creating two additional feature vectors representing different substructures of each sub-tree or neighborhood . The objective is then to predicting the existence/non-existence of a node in these newly defined features. To make the model capable of reasoning over long and short sequences, we will use an RNN with gated GGRU unit (Gated Gated Recurrent Unit) for sequence modeling. The goal is to learn from data that come in as inputs . The following are the steps: 1.) Create a set of nodes and edges within each graph. This means that you can train a model for unseen graphs as well. However, we aim to improve this by utilizing Graph Attention networks (GATs). For any new sub-graph it is possible by creating two additional feature vectors representing different parts or structures of the node pair in the neighborhood relationships . The objective is then to predicting the existence/non-existence of a node in these newly defined features. You want us to create an interactive web application that allows users to easily switch between various cursor themes for their desktop. Here are some specifications and guidelines: 1. You should consider color options, sizes, shapes, and types of the cursors. 40