This project shows how to setup LSTM based network for estimating bike rental demand and operationalize it.
- Execute every cell in the notebook
- Once all the cells are executed, operationalize using the following commands
Make sure the following packages are installed in the environment where this notebook is being executed:
- tensorflow
- h5py
- Keras
- azure-ml-api-sdk
Please use pip install to install above dependencies
Setup environment and publish web service. This assumes IT admin has already created modelmanagement account for you and have setup an ACS environment for you
#setup environment
$ az ml account modelmanagement set -n <model mgmt acct e.g. neerajteam2hosting> -g <azure resource group e.g. amlrsrcgrp2>
$ az ml env set -n <env name e.g. amlcluster> -g <azure resource group e.g. amlrsrcgrp2>
$ az ml env local
If you don't have previous environment or account setup, then please use the following commands to setup model management account and environment
$ az ml env setup -n <env name e.g. amlcluster> -g <azure resource group e.g. amlrsrcgrp2> # this will setup local environment. use option -c for cluster environment
$ az ml account modelmanagement create -n <acct name e.g. neerajteam2hosting> -l <location e.g. eastus2> -g <azure resource group e.g. amlrsrcgrp2> --sku-instances <sku count e.g. 1> --sku-name <pricing tier e.g. S1>
Publish web service using the following command by providing scoring file, model, runtime, dependency file, and schema
$ az ml service create realtime -f bikescore.py -m finalmodel.sav -r python -c ./aml_config/conda_dependencies.yml -s bikeschema.json -l true -n bikews1
Use the following command to list all active web services
$ az ml service list realtime -o table
$ az ml service usage realtime -i bikews1
Consume web service using the following command
$ az ml service run realtime -i bikews1 -d "{\"npa\": [[450.0, 353.0, 285.0, 332.0, 377.0, 268.0, 218.0, 387.0, 834.0, 508.0, 153.0, 42.0, 4.0, 1.0, 10.0, 17.0, 0.0, 4.0, 1.0, 2.0, 0.58, 0.5455, 0.64, 0.3284], [890.0, 450.0, 353.0, 285.0, 332.0, 377.0, 268.0, 218.0, 387.0, 834.0, 508.0, 153.0, 4.0, 1.0, 10.0, 18.0, 0.0, 4.0, 1.0, 2.0, 0.56, 0.5303, 0.64, 0.3284], [788.0, 890.0, 450.0, 353.0, 285.0, 332.0, 377.0, 268.0, 218.0, 387.0, 834.0, 508.0, 4.0, 1.0, 10.0, 19.0, 0.0, 4.0, 1.0, 2.0, 0.56, 0.5303, 0.68, 0.2985]]}"