I am working on object detection problem. i have train my model on Tensorflow, and got good detection accuracy. But now i have to deploy this on application like .Net technology. i have try TensorFlowSharp but it has many issues.
I need suggestion to further start work on which Deep learning platform that help me for the final product.
Caffe and Neon support easier platform switching (i.e., you can develop your model in one platform and deploy it on another). Deeplearning4j also has the capability to import models from several major frameworks including TensorFlow, Caffe, Torch and Theano.
@Mahdi, My Question is not switching between deeplearning platform such as Tensorflow, theano, etc.
i am working for a company, who want to integrate deep learning trained model with there .NET technology product.
I think Tensorflow Serving is a solution to your need. I have no experience with it but you may find useful information for using it at the following link:
https://medium.com/towards-data-science/how-to-deploy-machine-learning-models-with-tensorflow-part-1-make-your-model-ready-for-serving-776a14ec3198
- Badges
- Science topic