cpu_cores | small python library and utility to get the number

All the files in the same directory tree as entry_point will be packaged in the docker image created, along with the entry_point file.

Using the entry point, it puts the wheel into the tar and copies it to the bucket.

The "COPY" exists but comes after the "pip install" command, so pip does not see the wheel.

You have to patch tensorflow-cloud in containerize.py:

move the lines

If you want to create a custom environment you can use the below code to set the env configuration.

myenv = Environment(name="Environment")

myenv.docker.enabled = True

myenv.python.conda_dependencies = CondaDependencies.create(conda_packages = ['numpy','scikit-learn','pip','pandas'], pip_packages = ['azureml-defaults~= 1.34.0','azureml','azureml-core~= 1.34.0',"azureml-sdk",'inference-schema','azureml-telemetry~= 1.34.0','azureml- train-automl~= 1.34.0','azure-ml-api-sdk','python-dotenv','azureml-contrib-server','azureml-inference-server-http'])

Ref doc: https://docs.microsoft.com/en-us/python/api/azureml-core/azureml.core.environment(class)?view=azure-ml-py#:~:text=Upload%20the%20private%20pip%20wheel,in%20the%20workspace%20storage%20blob.&text=Build%20a%20Docker%20image%20for%20this%20environment%20in%20the%20cloud.&text=Build%20the%20local%20Docker%20or%20conda%20environment.

Please advise nginx docker image docs in the Using environment variables in nginx configuration section of the page.

The way the nginx docker image deals with environment variables is injecting them in runtime using the configs in the linked page

Without having the source of define_exp_parameters() it is impossible to exactly describe what goes wrong. But before calling that deployments is a list containing a single element that is a dict (with keys apiVersion, kind, etc.). And after that call deployments is a list of single elements list (which elements is aformentioned dict). You iterate over the "outer" list and dump a single element list, which, in block style, gives you the - which is the block sequence element indicator.

If you can't fix define_exp_parameters() to return a list for which each element is a dict again, you can just dump the first element of deployment:

Great to see people putting the R SDK through it's paces!

The vignette you're using is obviously a great way to get started. It seems you're almost all the way through without a hitch.

Deployment is always tricky, and I'm not expert myself. I'd point you to this guide on troubleshooting deployment locally. Similar functionality exists for the R SDK, namely: local_webservice_deployment_config().

So I think you change your example to this:

One concept that took me a while to get was the bifurcation of registering and using an Azure ML Environment. If you have already registered your env, myenv, and none of the details of the your environment have changed, there is no need re-register it with myenv.register(). You can simply get the already register env using Environment.get() like so:

I could solve my own problem by using python FileInput library.

You can use ProcessPoolExecutor from futures module in python.

An Executor subclass that executes calls asynchronously using a pool of at most max_workers processes. If max_workers is None or not given, it will default to the number of processors on the machine

you can find more at Python docs

Here, is the sample code of achieving the concurrency assuming that each msgList[i] is idependent of msgList[j] when i != j,