This is an automated email from the ASF dual-hosted git repository.

haibin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
     new 996d548  update README.md with details on typical usage of the scripts 
(#9108)
996d548 is described below

commit 996d54817d23d1e0d35fd70394ed09c9d45342e2
Author: Bhavin Thaker <bhavintha...@users.noreply.github.com>
AuthorDate: Wed Jan 3 14:34:50 2018 -0800

    update README.md with details on typical usage of the scripts (#9108)
---
 example/rnn/bucketing/README.md | 31 +++++++++++++++++++++++++++----
 1 file changed, 27 insertions(+), 4 deletions(-)

diff --git a/example/rnn/bucketing/README.md b/example/rnn/bucketing/README.md
index 6baf1ec..0481609 100644
--- a/example/rnn/bucketing/README.md
+++ b/example/rnn/bucketing/README.md
@@ -3,11 +3,34 @@ RNN Example
 This folder contains RNN examples using high level mxnet.rnn interface.
 
 ## Data
-Run `get_ptb_data.sh` to download PenTreeBank data.
+1) Review the license for the PenTreeBank dataset and ensure that you agree to 
it. Then uncomment the lines in the 'get_ptb_data.sh' script that download the 
dataset.
+
+2) Run `get_ptb_data.sh` to download PenTreeBank data.
 
 ## Python
 
-- [lstm_bucketing.py](lstm_bucketing.py) PennTreeBank language model by using 
LSTM
+- Generate the PennTreeBank language model by using LSTM:
+
+  For Python2 (CPU support): can take 2+ hours on AWS-EC2-p2.16xlarge
+
+      $ python  [lstm_bucketing.py](lstm_bucketing.py) 
+
+  For Python3 (CPU support): can take 2+ hours on AWS-EC2-p2.16xlarge
+
+      $ python3 [lstm_bucketing.py](lstm_bucketing.py) 
+
+  Assuming your machine has 4 GPUs and you want to use all the 4 GPUs:
+
+  For Python2 (GPU support only): can take 50+ minutes on AWS-EC2-p2.16xlarge
+
+      $ python  --gpus 0,1,2,3 
[cudnn_lstm_bucketing.py](cudnn_lstm_bucketing.py) 
+
+  For Python3 (GPU support only): can take 50+ minutes on AWS-EC2-p2.16xlarge
+
+      $ python3 --gpus 0,1,2,3 
[cudnn_lstm_bucketing.py](cudnn_lstm_bucketing.py) 
+
+
+### Performance Note:
+
+More ```MXNET_GPU_WORKER_NTHREADS``` may lead to better performance. For 
setting ```MXNET_GPU_WORKER_NTHREADS```, please refer to [Environment 
Variables](http://mxnet.incubator.apache.org/how_to/env_var.html).
 
-Performance Note:
-More ```MXNET_GPU_WORKER_NTHREADS``` may lead to better performance. For 
setting ```MXNET_GPU_WORKER_NTHREADS```, please refer to [Environment 
Variables](https://mxnet.readthedocs.org/en/latest/how_to/env_var.html).

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" <comm...@mxnet.apache.org>'].

Reply via email to