/Neural-Symbolic-Computing

Universität des Saarlandes - Neural Symbolic Computing - Semester Project - 2020

Primary LanguageJupyter Notebook

In this project, we will train a neural network on basic arithmetic operations on variables, which is a task that reoccurs very often while interpreting code.The goal of this project is to teach a neural network a basic arithmetic understanding of addition and subtraction of two variables x and y.

Neural-Symbolic-Computing

model_params.pyt contains the saved model, will be automatically downloaded if LOAD under Testis set to True. In case one wants to just load the model, just execute all cells except the ones under Training and Loss progression then use the cells under Test to test the model for different inputs.

to train and test the trained model, please set LOAD=False under Test for testing the (newly) trained model.

You can download model_params.pyt from this link: https://oc.cs.uni-saarland.de/owncloud/index.php/s/NN5N2jGT6fktKdd