function-approximation Notebooks of simple models to comprehend machine learning with pytorch. function approximation sin Encoding 64b input to 32b methods and techniques quantization - 32 bit weights to 8 bit