Can HumDay connect to Google Cloud AutoML?

Yes. HumDay reads Google Cloud AutoML's own published API description and derives the operations from it, so there is no hand-built connector to wait for. 25 operations are documented.

Do I need to write code to automate Google Cloud AutoML?

No. You describe the outcome you want in plain words. HumDay agrees a contract with you, writes the program, and shows you a test run before anything touches your Google Cloud AutoML account.

What can HumDay do in Google Cloud AutoML?

16 of the 25 documented operations change something in Google Cloud AutoML, and 9 read from it. HumDay only ever uses the specific operations your approved contract needs.

Is my Google Cloud AutoML account safe?

Your credentials are stored encrypted and are never shown in chat, code, or logs. Every run is limited to the operations you explicitly approved, and anything that writes to Google Cloud AutoML is held behind that approval.

Automate Google Cloud AutoML

HumDay reads Google Cloud AutoML’s own published API description and derives 25 operations from it. Describe the outcome you want in plain words — you get a program that is written, proven on real data, and run for you.

What Google Cloud AutoML is

Train high-quality custom machine learning models with minimum effort and machine learning expertise.

API host: automl.googleapis.com

What HumDay can do in Google Cloud AutoML

16 documented operations change something in Google Cloud AutoML.

POSTCreates a model. Returns a Model in the response field when it completes. When you create a model, several model evaluations are created for it: a global evaluation, and one evalu…/v1beta1/{parent}/models
POSTExports a trained, "export-able", model to a user specified Google Cloud Storage location./v1beta1/{name}:export
POSTCreates a dataset./v1beta1/{parent}/datasets
DELETEDeletes a long-running operation./v1beta1/{name}
PATCHUpdates a column spec./v1beta1/{name}
POSTWaits until the specified long-running operation is done or reaches at most a specified timeout, returning the latest state./v1beta1/{name}:wait
POSTStarts asynchronous cancellation on a long-running operation./v1beta1/{name}:cancel
POSTDeploys a model. If a model is already deployed, deploying it with the same parameters has no effect. Deploying with different parametrs (as e.g. changing nodenumber) will reset t…/v1beta1/{name}:deploy
POSTPerform an online prediction./v1beta1/{name}:predict
POSTUndeploys a model. If the model is not deployed this method has no effect. Only applicable for Text Classification, Image Object Detection and Tables; all other domains manage dep…/v1beta1/{name}:undeploy
POSTExports dataset's data to the provided output location./v1beta1/{name}:exportData
POSTImports data into a dataset./v1beta1/{name}:importData
POSTPerform a batch prediction./v1beta1/{name}:batchPredict
POSTSets the access control policy on the specified resource./v1beta1/{resource}:setIamPolicy
POSTReturns permissions that a caller has on the specified resource./v1beta1/{resource}:testIamPermissions
POSTExports examples on which the model was evaluated (i.e./v1beta1/{name}:exportEvaluatedExamples

What HumDay can read from Google Cloud AutoML

These are the operations a schedule or a trigger can watch.

GETGets the latest state of a long-running operation./v1beta1/{name}
GETGets the access control policy for a resource./v1beta1/{resource}:getIamPolicy
GETLists models./v1beta1/{parent}/models
GETLists information about the supported locations for this service./v1beta1/{name}/locations
GETLists operations that match the specified filter in the request./v1beta1/{name}/operations
GETLists datasets in a project./v1beta1/{parent}/datasets
GETLists table specs in a dataset./v1beta1/{parent}/tableSpecs
GETLists column specs in a table spec./v1beta1/{parent}/columnSpecs
GETLists model evaluations./v1beta1/{parent}/modelEvaluations

How automating Google Cloud AutoML works

Describe the outcome. Say what you want to happen, in your own words. No node graphs, no field mapping.
Approve the contract. HumDay writes down exactly what it will do, what it will touch, and what it will never do. You approve it before anything is built.
See it proven. The program runs and shows you the result before it is allowed near your live Google Cloud AutoML account.
Grant access, then go live. You approve the specific Google Cloud AutoML operations it may use — and only those.