Gcloud ML-引擎预测错误OOM 429 [英] Gcloud ML-Engine Prection Error OOM 429

问题描述

尝试使用gcloud ml-engine predict

ERROR: (gcloud.ml-engine.predict) HTTP request failed. Response: {
  "error": {
    "code": 429,
    "message": "Prediction server is out of memory, possibly because model size is too big.",
    "status": "RESOURCE_EXHAUSTED"
  }
}

我的模型大小为151 mb，我也在使用Tensorflow 1.4版，该版本不需要variables文件夹.执行预测时，它会使用超过2gb的内存.我使用的是初始版本的修改版本.

My model size is 151 mb, I'm also using Tensorflow version 1.4 that does not requiere variables folder. When performing prediction it uses over 2gb. I'm using a modified version of inception.

推荐答案

当前，用于预测的计算机只有2 GB的RAM.我们正在努力将具有更多RAM的机器引入该服务.

Currently, the machines used for prediction have only 2 GB of RAM. We are working on bringing machines with more RAM to the service.

也就是说，原始"初始模型在磁盘上的大小通常与您所报告的大小相同，但往往很容易装入2 GB的RAM.考虑到您所做的更改，RAM会爆炸吗?

That said, "vanilla" inception models are usually about the same size on disk as you are reporting but tend to easily fit in 2 GB of RAM. Is the explosion of RAM expected given the changes that you've made?

这篇关于Gcloud ML-引擎预测错误OOM 429的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！