Error in mx.nd.internal.empty.array(shape, ctx) : [14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enable GPU usage in R

[jasperDD]

I run this script for captcha recognition in R (windows machine)
mx.set.seed(42)

model <- mx.model.FeedForward.create(
X = train,
eval.data = val,
ctx = mx.gpu(),
symbol = captcha_net,
eval.metric = mx.metric.acc2,
num.round = 10,
learning.rate = 0.0001,
momentum = 0.9,
wd = 0.00001,
batch.end.callback = mx.callback.log.train.metric(50),
initializer = mx.init.Xavier(factor_type = "in", magnitude = 2.34),
optimizer = "sgd",
clip_gradient = 10
)

Then, it returns error
[14:58:38] c:\jenkins\workspace\mxnet\mxnet\src\operator\tensor./matrix_op-inl.h:166: Using target_shape will be deprecated.
Start training with 1 devices
[14:58:38] c:\jenkins\workspace\mxnet\mxnet\src\operator\tensor./matrix_op-inl.h:166: Using target_shape will be deprecated.
Error in mx.nd.internal.empty.array(shape, ctx) :
[14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enable GPU usage

What;s wrong and how to fix it?

[jasperDD]

I understood what the problem. I don't have Nvidia card, so i must use MXnet only CPU and not GPU
like this
model <- mx.model.FeedForward.create(
X = train,
eval.data = val,
ctx = mx.cpu(),
symbol = captcha_net,
eval.metric = mx.metric.acc2,
num.round = 10,
learning.rate = 0.0001,
momentum = 0.9,
wd = 0.00001,
batch.end.callback = mx.callback.log.train.metric(50),
initializer = mx.init.Xavier(factor_type = "in", magnitude = 2.34),
optimizer = "sgd",
clip_gradient = 10
)

[lanking520]

@jasperDD Thanks for identifying the problem. Can you confirm that the model is using GPU as a default?
@anirudhacharya @hetong007

[Ishitori]

Yes, @lanking520, it is using GPU by default. See it here: https://github.com/apache/incubator-mxnet/blob/master/example/captcha/mxnet_captcha.R

[lanking520]

@Ishitori Thanks for your input. Should we make it cpu as a default?