can-lehmann/exprgrad

This is stars and forks stats for /can-lehmann/exprgrad repository. As of 03 May, 2024 this repository has 109 stars and 0 forks.

Exprgrad Exprgrad is an experimental deep learning framework for Nim based on a differentiable array programming language. Exprgrad makes creating and training neural networks easy: import std/random import exprgrad, exprgrad/layers/[base, dnn] randomize(10) let net = input("x") .dense(2, 4).leakyRelu() # 1st Layer .dense(4, 1).sigmoid() # 2nd Layer .target("predict") .mse(input("y")) # Loss .target("loss") .backprop(gradientDescent.makeOpt(rate=0.1)) # Train ...