Skip to Content

ML Foundation Bring your own Model


I have a Tensorflow saved Model in .pb Format which is packed in a .zip and want to deploy it in my SCP Cloud with Cloud Foundry Trial.

There is a Blog Article and a Documentation which are a good help.

But when I am deploying my model and check the status of the Model after a long Pending status I get:

"deploymentStatus": {
                "state": "FAILED",
                "description": "Deployment could not be completed due to a timeout. Please try again after some time or contact administrator"

Well that's really hard to debug, without any other Information, so maybe someone also got experience with it?

Add comment
10|10000 characters needed characters exceeded

  • Follow
  • Get RSS Feed

3 Answers

  • Best Answer
    Jun 18, 2018 at 07:47 AM

    I solved the Problem,

    the deployed Model has to be in the savedModel Format, and needs to be zipped. I just tried it with a .pb file which is not enough.

    If you're using Keras to create the model, use this code here to transform the model in the right format and the deployment should work fine.

    Add comment
    10|10000 characters needed characters exceeded

  • Jun 14, 2018 at 03:12 PM

    Hi Benny Mueller,

    i try to check this, but first of all which api version did u use v1 or v2?



    Add comment
    10|10000 characters needed characters exceeded

  • Jun 14, 2018 at 03:28 PM

    Hi Benny Mueller,

    i´ve tried it now successfully with v2.

    The following steps are required:

    1. Download the "" from the SAP Help here

    2. Add the zip archive to the repository by executing this request:

    Method: POST

    URL: https://*

    for my i´ve named the model "my-model", the url looks now like this:


    After a while i got this response:


    "namespace": "8dd19b31-7dc5-4620-be78-8d5ebbfa5d61",

    "modelName": "my-model",

    "version": "1", "metaData": null,

    "checkSum": null,

    "createdAt": "2018-06-14T15:17:11.569+0000",

    "updatedAt": "2018-06-14T15:17:11.569+0000",

    "fileName": ""


    3. Deploy the model:

    Method: POST

    URL: https://*

    Body: {"modelName": "my-model"}

    4. Some minutes later i execute the following GET request:

    URL: https://*


    hopefully this helps!?

    best regards,


    Add comment
    10|10000 characters needed characters exceeded

    • Hello Fabian,

      Thanks for your interest, I did all steps succesfully until the Step 3: Deploy the model. When I tried to POST model via Postman. I got the 404 Not Found Error. I guess it's related to trial account, Are you sure, these steps includes Trial Account?

      Thanks, have a nice coding :)

      Edit : Actually, I did it using different host. "DEPLOYMENT_API_URL" instead of "MODEL_REPO_URL". When I checked in Swagger API. I faced new problem :). It is deprecated. I am confused, What I am doing?

      deprecated.png (57.5 kB)