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
From the paper:
"we experimented with initializing the hidden states with zeros on half of the examples in the batch, and with standard Gaussian noise on the rest of the examples"
"Mixed initialization: During each training forward pass, each sample was assigned with either zero initialization (i.e. the fixed point was initialized with the 0 vector) or standard normal distribution (i.e. ...) using a Bernoulli random variable of probability 0.5 (i.e. the examples that were run with zero vs. normal initializations were roughly half-half."
This form has the disadvantage of assuming that all but the last dimension are batch dimensions. But this seems to be quite a reasonable assumption, and downstream users can easily adjust to this by reshaping and rearranging the dimensions.
The text was updated successfully, but these errors were encountered:
From the paper:
"we experimented with initializing the hidden states with zeros on half of the examples in the batch, and with standard Gaussian noise on the rest of the examples"
"Mixed initialization: During each training forward pass, each sample was assigned with either zero initialization (i.e. the fixed point was initialized with the 0 vector) or standard normal distribution (i.e. ...) using a Bernoulli random variable of probability 0.5 (i.e. the examples that were run with zero vs. normal initializations were roughly half-half."
Current implementation:
torchdeq/torchdeq/utils/init.py
Lines 4 to 21 in 4f6bd5f
It seems more appropriate to do this instead to match the paper.
This form has the disadvantage of assuming that all but the last dimension are batch dimensions. But this seems to be quite a reasonable assumption, and downstream users can easily adjust to this by reshaping and rearranging the dimensions.
The text was updated successfully, but these errors were encountered: