EasySRL icon indicating copy to clipboard operation
EasySRL copied to clipboard

Running LSTM parser on GPU

Open HGrant15 opened this issue 7 years ago • 2 comments

Hi,

I am trying to run your parser over the trained model provided. I would like to run in on my GPU, however it always runs on the CPU.

Do I need to make any changes to the build.xml to compile the project with the tensorflow.jar and libtensorflow.so files?

I noticed in the codebase that the GPUEvaluation.java makes use of TensorFlowInputReader which the rest of the codebase does not use. Is this class the only part of the codebase that makes use of the GPU?

I realise this would have been more appropriate as an email question rather than an 'issue' but I can not close this topic.

Thanks.

HGrant15 avatar Nov 25 '17 09:11 HGrant15

The GPU version can be used with two simple changes: (1) Use this model instead. (2) Download this binary and place it in the lib directory.

An example of a script preparing this (amongst other things) can be found here.

Hope that helps!

kentonl avatar Nov 29 '17 19:11 kentonl

java -jar easysrl.jar --model modelFolder --outputFormat ccgbank does not work

arunikayadav42 avatar Nov 19 '19 07:11 arunikayadav42