This is a simple recreation of torch and I will see if I can use it to train a model

This is a replica of Tensors in torch, and has multiple funcitons. Included in them are funcitons like ad, multiply, div, tanh, sqrt

Has linear, Embedding, LayerNorm, Relu, Sequential, SGD/Adam, CrossEntropyLoss, MSELoss, and softmax

Transformer with Self-Attention

Example in which you can train the model to predict number sequences

Have fun!