-
Notifications
You must be signed in to change notification settings - Fork 5.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
run vgg_16_cifar wrong. #63
Comments
Maybe Out of Memory, See in /var/log/messages |
It says that
We will give a documentation to guide how to use less memory to train the job. |
遇到了相同的问题。。 |
@zhangscth https://github.com/baidu/Paddle/pull/128/files#diff-d718076d937b4f0e340765bf95c122c3 如果是killed的话,看一下如何减少内存占用。这块文档还在写 |
@zhangscth @zhuyong0000 如何减少内存的文档:http://www.paddlepaddle.org/doc_cn/faq/index.html#id1 在这个demo中您可以尝试减少DataProvider缓存、或者减少batch size. |
Fix unittest for predict and evaluate
refactor Object.As to as to remind it lack type check
* add InstanceNorm and LayerNorm Ops * add annotation * transpose * merge develop-ipu * merge develop-ipu * add GetOutputNode * delete layernorm from backend * groupnorm, instancenorm , layernorm * transpose, reshape * reshape, transpose unitest * pre-commit
* Update express_ner example * update run_bigru_crf * fix msra_ner example
add download lib doc
* Update README * Change docs dir to en_US and zh_CN * Add paper reference * Fix link
expand slot's feasign for cvr model in fused_cvm_op
Implement iterator vars fetching in ReduceOp
xpu support check_nan_inf
deploy add README.md
运行./train.sh 出现下面错误:$MYDIR/../opt/paddle/bin/paddle_trainer $ {@:2}
I0910 08:11:05.670004 1881 GradientMachine.cpp:134] Initing parameters..
I0910 08:11:06.422868 1881 GradientMachine.cpp:141] Init parameters done.
/usr/local/bin/paddle: line 46: 1881 Killed ${DEBUGGER}
No data to plot. Exiting!
运行的环境是paddledev/paddle:cpu-demo-latest,paddle版本为:
PaddlePaddle 0.8.0b, compiled with
with_avx: OFF
with_gpu: OFF
with_double: OFF
with_python: ON
with_rdma: OFF
with_glog: ON
with_gflags: ON
with_metric_learning:
with_timer: OFF
with_predict_sdk:
The text was updated successfully, but these errors were encountered: