[
https://issues.apache.org/jira/browse/SINGA-339?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
wentong updated SINGA-339:
--------------------------
Environment:
Python v2.7.13;
Singa: singa-cuda7.5-cudnn5 v1.1.1;
Server: singa1
was:Python v2.7.13; singa-cuda7.5-cudnn5 v1.1.1; server: singa1
> Char-RNN fails to run on sample text
> ------------------------------------
>
> Key: SINGA-339
> URL: https://issues.apache.org/jira/browse/SINGA-339
> Project: Singa
> Issue Type: Bug
> Environment: Python v2.7.13;
> Singa: singa-cuda7.5-cudnn5 v1.1.1;
> Server: singa1
> Reporter: wentong
> Priority: Minor
>
> When I tried to run the char-rnn example in singa with the provided sample
> text "linux_input.txt", I got the following error message:
> wentong@singa1:~/incubator-singa/examples/char-rnn$ python train.py
> "linux_input.txt"
> train dat (76393, 65)
> val dat (19099, 65)
> created rnn
> rnn weight l1 = 0.040077
> dense w (128, 101)
> dense b (101,)
> [....................] 100.0% trainingEpoch 0, train loss is
> 2.486698123326533273295897369690474792630041774822640536023023408687341856910WARNING:
> Logging before InitGoogleLogging() is written to
> STDERR339175394116557228656463428434357538F1031 13:53:31.595016 24408
> dense.cc:54] Check failed: buf_.empty()
> *** Check failure stack trace: ***
> Aborted (core dumped)
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
