Top Questions

1. Will all the variables be updated (through backprop+optim) w/o using tf.stop_gradient in tensorflow2.0/1.0 or with stop_gradient=False in paddlepaddle
   1 votes

   0 answers