from caffe2.python import workspace, model_helper, brew
import numpy as np
x = np.ones((4, 100, 100, 3)).astype(np.float32)
m = model_helper.ModelHelper(name='train')
conv1 = brew.conv(m, 'x', 'conv_1', 3, 5, 3)
for i in range(0, 2):
why this is throwing error ?
Could you please share the error so that we might get a better insight to your problem.
WARNING:root:This caffe2 python run does not have GPU support. Will run in CPU only mode.
WARNING: Logging before InitGoogleLogging() is written to STDERR
W1214 14:06:30.382432 165639616 init.h:99] Caffe2 GlobalInit should be run before any other API calls.
W1214 14:06:30.382798 165639616 init.h:99] Caffe2 GlobalInit should be run before any other API calls.
I1214 14:06:30.383255 165639616 operator.cc:169] Engine CUDNN is not available for operator Conv.
From the error it seems the script is trying to use Cuda but the support is not available. Is cuda installed properly? Do you have GPU support?
Thanks for the reply. I don’t have CUDA and I get that part. I don’t understand the global init error. Why that’s happening?