You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2019/06/13 03:19:27 UTC
[GitHub] [incubator-mxnet] chinakook opened a new issue #15231:
mxnet.nd.random.multinomial is very slow on CPU and stucked on GPU
chinakook opened a new issue #15231: mxnet.nd.random.multinomial is very slow on CPU and stucked on GPU
URL: https://github.com/apache/incubator-mxnet/issues/15231
```python
import time
import numpy as np
import mxnet as mx
p = [1 / 300000.] * 300000
probs = np.array(p)
t0 = time.time()
sample_times = np.random.multinomial(len(probs), np.array(probs))
t1 = time.time()
print("Numpy MultiNomial Time", t1 - t0)
probs = mx.nd.array(p, ctx=mx.cpu())
t0 = time.time()
indices = mx.nd.random.multinomial(probs, len(probs))
mx.nd.waitall()
t1 = time.time()
print("MXNet CPU MultiNomial Time", t1 - t0)
# stucked
# probs = mx.nd.array(p, ctx=mx.gpu(0))
# t0 = time.time()
# indices = mx.nd.random.multinomial(probs, len(probs))
# mx.nd.waitall()
# t1 = time.time()
# print("MXNet GPU MultiNomial Time", t1 - t0)
```
The result is
```
Numpy MultiNomial Time 0.01881909
MXNet GPU MultiNomial Time 43.93881797
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services