We are doing the following Trained with GPUs using the GPU bindings. Serialize that model to disk. Using the CPU bindings we deserialize and get an error (see below) Is it possible to support this use case and is the GPU really required for scoring? [main] INFO org.nd4j.linalg.factory.Nd4jBacke…

GPU training, CPU scoring?

dawood June 29, 2020, 1:09pm 3

Sorry, but I guess I posted this prematurely. The person seeing this can not reproduce. Thanks.

Topic		Replies	Views
GPU Issue, memory not freed DL4J	3	501	May 7, 2020
Only two threads are running when switching training to GPU DL4J	2	419	May 29, 2020
LSTMHelper error with GPU backend DL4J	2	572	May 23, 2021
Model works with CPU, but outputs all 0's with CUDA DL4J	8	432	April 1, 2021
25% GPU Usage on 1080 ti DL4J	9	585	February 14, 2022