You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I believe it's just wrong assert. The assert is after input assignment, so ".d" derivatives parts are still un-initialized. (for larger networks you just increase probability that there will be random nan).
This PR solves the NaN issue reported here : tmbdev#63
The problem is that for newly allocated batches inside a Sequence, memory had to be reset to zeros, otherwise NaNs may flow in
For many models (especially deep ones with many parameters e.g. bidi2), I keep getting the following error
clstm.cc:664: void ocropus::GenericNPLSTM<F, G, H>::backward() [with int F = 1; int G = 2; int H = 2]: Assertion `!anynan(out)' failed.
where the old version (Mat-based) works just fine
The text was updated successfully, but these errors were encountered: