l***@gmail.com
2017-11-25 00:57:00 UTC
Hello everyone,
Recently, I use theano (0.9.0) and pygpu (0.6.9) to conduct some
experiments and I found that different results are obtained if I downgrade
the version of pygpu to 0.6.2. I think such difference does not come from
the random initialization of the neural network because there always exists
1-2% performance gap. Thus, I wonder if the reason lies in the version of
pygpu.
Have you ever encountered this problem?
Recently, I use theano (0.9.0) and pygpu (0.6.9) to conduct some
experiments and I found that different results are obtained if I downgrade
the version of pygpu to 0.6.2. I think such difference does not come from
the random initialization of the neural network because there always exists
1-2% performance gap. Thus, I wonder if the reason lies in the version of
pygpu.
Have you ever encountered this problem?
--
---
You received this message because you are subscribed to the Google Groups "theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to theano-users+***@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
---
You received this message because you are subscribed to the Google Groups "theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to theano-users+***@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.