[英]Deploying a Tensorflow/Keras model in Spark Pipeline
I have trained several RNN+biLSTM
models that I want to deploy in a pipeline consisting of pyspark
pipeline steps. 我已经训练了几种
RNN+biLSTM
模型,我希望将它们部署在包含pyspark
管道步骤的管道中。 spark-deep-learning
seems to be a stale project that only accommodates work with image data. spark-deep-learning
似乎是一个过时的项目,只能容纳图像数据。 Are there any best practices today for loading tensorflow
/ keras
models (and their associated vector embeddings) into pyspark
pipelines? 今天是否有将
tensorflow
/ keras
模型(及其相关的矢量嵌入) pyspark
到pyspark
管道中的最佳实践?
If you want to deploy a tensorflow model into Spark, you should take a look at Deeplearning4J . 如果要将张量流模型部署到Spark中,则应查看Deeplearning4J 。 It comes with some Importers, where you can read keras and TensorFlow models.
它带有一些导入程序,您可以在其中读取keras和TensorFlow模型。 Be aware, that not every layer is supported.
请注意,并非每层都受支持。
Besides spark-deep-learning there is tensorframe , i never used it , so I don´t know how good it is. 除了火花深度学习之外 ,我还没有使用过tensorframe ,所以我不知道它有多好。
In general I would suggest to use tensorflow directly via Distributed Tensorflow and not using all these wrappers. 通常,我建议直接通过分布式Tensorflow使用tensorflow,而不要使用所有这些包装器。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.