- How to build a serving model in Tensorflow ?
- How to use estimator effectively ?
- Serving the model in cloud ml
- Using Keras with Tensorflow Estimator
- How to serve scikit-learn models in Cloud-ML(WIP) ?
- Many More.......
The idea is to solve many real-world data problems using Machine Learning with Tensorflow. We will use various open dataset available from kaggle and other sources to solve the required problem using Tensorflow, scikit and deploying it in cloud ml.
This repository also contains implementation of research papers in tensorflow which can be deployed in cloud-ml
Feel free to support this repository by giving Star.
Contributions are welcome !!
To deploy the model in cloud-ml use the following command
gcloud ml-engine versions create {MODEL_VERSION} --model={MODEL_NAME} --origin={MODEL_PATH_IN_BUCKET} --runtime-version=1.4
gcloud ml-engine versions create {MODEL_VERSION} --model={MODEL_NAME} --origin={MODEL_PATH(PICKLED FILE)} --runtime-version="1.4" --framework="SCIKIT_LEARN"
gcloud ml-engine predict --model={MODEL_NAME} --version={MODEL_VERSION} --json-instances={INPUT_JSON_FILE}