nickgreenquist / cu2rec

CUDA Implementation of Parallel Matrix Factorization Algorithm for Recommender Systems
GNU Affero General Public License v3.0
13 stars 1 forks source link

compute single loss in sgd, move prediction code to devicde function,… #31

Closed nickgreenquist closed 5 years ago

nickgreenquist commented 5 years ago

… add biases to shared mem, add timers to training loop and total loss calcs