Anton Fedotov
Results
26
issues of
Anton Fedotov
For solving this issue these steps needed: * Add method `enable_grads_acumulation(steps_num: int)` to `Trainer` class * Add gradients accumulating like described [there](https://discuss.pytorch.org/t/how-to-implement-accumulated-gradient-in-pytorch-i-e-iter-size-in-caffe-prototxt/2522/3) * Write tests for a simple network, there...
good first issue
The idea is cpu and gpu occupancy monitoring for analyse code pervormance