I build sequential model with keras and tensorflow backend. As you seen on picture, model uses all cores but none of them 100%. I would like to know, how to make keras+tf utilize every core by 100%.
I tried to build tensorflow from scratch and reinstall, but it didn't help.
It turned out, my model was too simple to utilize 100% cpu. After adding more neurons to the network, cpu utilization increased to almost 100%.
I dont know, why keras don't use availible cpu power to compute simple models faster.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With