CoolMind


If you cannot measure it, you cannot improve it.

Estimator core in evaluator

记录另外一次 Estimator 的 core 排查。 背景 复现代码 Lessons learned 参考 背景 我们用户使用 Estimator 都是使用 chief + worker + ps + evaluator 的形式。在线上偶发的 evaluator 这...

Estimator OOM after export model frequently

最近用户反馈一个 Estimator 的问题,OOM (out of memory)。由于完全使用的是 Estimator 的逻辑,出错问题 排查了很久。最终使用多个工具找到问题的根源。ps_util 打印内存,objgraph 寻找内存泄露。 背景 序列化问题 模型导出 ...

Feature column a tutorial

特征列 feature column 深度神经网络输入 Feature columns 引入 Tensorflow 和其它库 构造示例数据 示例数据 演示几种不同类型的特征列 数值列 (Numeric columns) 分桶列 (Bucketized co...

Estimator a tutorial

介绍 Estiamtor 相关知识。 背景 非 Estimator 基于 Parameter server 架构的分布式学习 Estimator 行为控制 构造 Estimator 使用创建的 Estimator ...

Python unittest and unittest.mock

Each language has its own known testing framework which makes writing test cases easier. In Python, the standard and built-in are unittest module. ...

Tensorflow save and restore A tutorial

Introduction Show me the code Checkpoint 格式 SavedModel Directory structure Cpp CheckpointReader implemen...

Running Median

Problem Solution Problem This problem was asked by Microsoft. Compute the running median of a sequence of numbers. That is, given a stream...

Run-length encoding

Problem Solution Problem This problem was asked by Amazon. Run-length encoding is a fast and simple method of encoding strings. The basic ...