Merge pull request #28 from deephdc/ignacio-br0

Clean images and add serverless
deephdc · Apr 15, 2020 · 2320f37 · 2320f37
2 parents 6aa8963 + 991ca77
commit 2320f37
Show file tree

Hide file tree

Showing 11 changed files with 41 additions and 4 deletions.
diff --git a/source/_static/dashboard-configure.png b/source/_static/dashboard-configure.png
diff --git a/source/_static/dashboard-deployments.png b/source/_static/dashboard-deployments.png
diff --git a/source/_static/dashboard-history-full.png b/source/_static/dashboard-history-full.png
diff --git a/source/_static/dashboard-history.png b/source/_static/dashboard-history.png
diff --git a/source/_static/dashboard-home.png b/source/_static/dashboard-home.png
diff --git a/source/_static/paas-dashboard.png b/source/_static/paas-dashboard.png
diff --git a/source/index.rst b/source/index.rst
@@ -36,6 +36,7 @@ following section.
     * `DEEP IAM <https://iam.deep-hybrid-datacloud.eu/>`_
     * `DEEP Open Catalog - Marketplace <https://marketplace.deep-hybrid-datacloud.eu/>`_
     * DEEP Dashboard (`Training <https://train.deep-hybrid-datacloud.eu/>`_, `Advanced <https://paas.cloud.cnaf.infn.it/>`_)
+    * `DEEPaaS: DEEP as a Service <https://deepaas.deep-hybrid-datacloud.eu/>`_
     * `DEEP Nextcloud <https://nc.deep-hybrid-datacloud.eu>`_
     * `Official GitHub <https://github.com/deephdc>`_
     * `Official DockerHub <https://hub.docker.com/u/deephdc/>`_

diff --git a/source/user/howto/index.rst b/source/user/howto/index.rst
@@ -10,6 +10,7 @@ HowTo's
    Train a model locally <train-model-locally>
    Train a model remotely <train-model-remotely>
    Perform inference locally <inference-locally>
+   Perform inference remotely <inference-remotely>
    Add module to the DEEP marketplace <add-to-DEEP-marketplace>
    Use rclone <rclone>
    Install and configure oidc-agent <oidc-agent>

diff --git a/source/user/howto/inference-remotely.rst b/source/user/howto/inference-remotely.rst
@@ -0,0 +1,23 @@
+.. highlight:: console
+
+**********************
+Try a service remotely
+**********************
+
+
+The first step is to go to the `DEEP as a Service (or DEEPaaS) <https://deepaas.deep-hybrid-datacloud.eu/>`_.
+For educational purposes we are going to use a `general model to identify images <https://marketplace.deep-hybrid-datacloud.eu/modules/train-an-image-classifier.html>`_. This will allow us to see the general workflow.
+
+Go to ``Swagger Interfaces`` and click ``Swagger UI for "image-classification-tf"``.
+You will see the API documentation, where you can test the module's functionality, as well as perform other actions.
+
+.. image:: ../../_static/deepaas.png
+  :width: 500
+
+Go to the  ``predict()`` function and upload the file/data you want to predict (in the case of the image classifier
+this should be an image file). The appropriate data formats of the files you have to upload are often discussed
+in the module's Marketplace page.
+
+The response from the ``predict()`` function will vary from module to module but usually consists on a JSON dict
+with the predictions. For example the image classifier return a list of predicted classes along with predicted accuracy.
+Other modules might return files instead of a JSON.
diff --git a/source/user/overview/architecture.rst b/source/user/overview/architecture.rst
@@ -29,8 +29,19 @@ general image classifier.
 For more information have a look at the `marketplace <https://marketplace.deep-hybrid-datacloud.eu/>`_.
 
 
-The API
-=======
+DEEP as a Service
+=================
+
+`DEEP as a Service (or DEEPaaS) <https://deepaas.deep-hybrid-datacloud.eu/>`_ is a fully managed service that allows
+to easily and automatically deploy developed applications as services, with horizontal scalability thanks to a
+serverless approach. Module owners only need to care about the application development process, and incorporate
+new features that the automation system receives as an input.  The serverless framework  allows any user to
+automatically deploy from the browser any module in real time to try it. We only allow trying prediction, for training,
+which is more resource consuming, users must use the DEEP Training Dashboard.
+
+
+The DEEPaaS API
+===============
 
 The :doc:`DEEPaaS API <api>` is a key component for making the modules accessible to everybody (including non-experts), as it
 provides a consistent and easy to use way to access the model's functionality. It is available for both inference and training.

diff --git a/source/user/overview/user-roles.rst b/source/user/overview/user-roles.rst
@@ -18,12 +18,13 @@ that has been containerized, and use it to classify their own plant images.
 **What DEEP can offer to you:**
 
 * a :ref:`catalogue <user/overview/architecture:The marketplace>` full of ready-to-use modules to perform inference with your data
-* an :ref:`API <user/overview/architecture:The API>` to easily interact with the services
+* an :ref:`API <user/overview/architecture:The DEEPaaS API>` to easily interact with the services
 * solutions to run the inference in the Cloud or in your local resources
 * the ability to develop complex topologies by composing different modules
 
 **Related HowTo's:**
 
+* :doc:`How to perform inference remotely <../howto/inference-remotely>`
 * :doc:`How to perform inference locally <../howto/inference-locally>`
 
 
@@ -42,7 +43,7 @@ model and retrains it to perform `seed classification <https://marketplace.deep-
 **What DEEP can offer to you:**
 
 * the ability to train out-of-the-box a module of the :ref:`catalogue <user/overview/architecture:The marketplace>` on your personal dataset
-* an :ref:`API <user/overview/architecture:The API>` to easily interact with the model
+* an :ref:`API <user/overview/architecture:The DEEPaaS API>` to easily interact with the model
 * data :ref:`storage resources <user/overview/architecture:The data storage resources>` to access your dataset
   (`DEEP-Nextcloud <https://nc.deep-hybrid-datacloud.eu>`_, `OneData <https://onedata.org/>`_, ...)
 * the ability to deploy the developed service on Cloud resources