Open feina1 opened 6 years ago
The C++ implementation has all the code to do ASGD training, but there always was a small loss in final training accuracy, so we never fully cleaned up and released the scripts, and focused on the Tensorflow implementation instead. We are currently investigating parallelization within this framework, using Horovod.
On Dec 26, 2017, at 6:47 AM, ramonsanabria notifications@github.com wrote:
Hi,
It is in our TODO list. But we do not have any estimate yet.
Thanks!
On Tue, Dec 26, 2017 at 12:30 PM feina1 notifications@github.com wrote:
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/srvk/eesen/issues/165, or mute the thread https://github.com/notifications/unsubscribe-auth/AMlwPfhI3ONntw7g1P40igA7zCawHw3Tks5tENjhgaJpZM4RMtTs .
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/srvk/eesen/issues/165#issuecomment-353959652, or mute the thread https://github.com/notifications/unsubscribe-auth/AEnA8WEjZgDOWZHSyHeJvW7L8c-jGs2Cks5tENzGgaJpZM4RMtTs.
Florian Metze http://www.cs.cmu.edu/directory/florian-metze Associate Research Professor Carnegie Mellon University
Hi,
It is in our TODO list. But we do not have any estimate yet.
Thanks!
On Tue, Dec 26, 2017 at 12:30 PM feina1 notifications@github.com wrote: