Feature/SK-728 | Move Studio docs to Fedn docs #555
Conversation
docs/studio.rst
Outdated
| Studio | ||
| =============== | ||
|
|
||
| FEDn Studio is a web-based tool for managing and monitoring federated learning experiments. It provides a user-friendly interface for creating and managing federated learning experiments, monitoring the progress of training, and visualizing the results. |
There was a problem hiding this comment.
Repetition in text. Also highlight that Studio provides additional security (token-based auth, RBAC) and a managed FEDn network.
docs/studio.rst
Outdated
| Create a project | ||
| ---------------- | ||
|
|
||
| Start by creating a new project. A project is a container for your federated learning experiments. You can create multiple projects to organize your experiments. A project can be shared with other users, allowing you to collaborate on experiments. |
There was a problem hiding this comment.
I think avoid 'container' to minimize risk that someone thinks Docker.
docs/studio.rst
Outdated
| Configure and attach clients | ||
| ---------------------------- | ||
|
|
||
| This step is very similar to how connecting a client to a local FEDn instance (:doc:`quickstart`). The main difference is that the client needs an authorization token to connect to a federation hosted on FEDn Studio. These are the steps to configure and attach a client to a project: |
There was a problem hiding this comment.
..to how a client is connected to a local FEDn....
docs/studio.rst
Outdated
|
|
||
| This step is very similar to how connecting a client to a local FEDn instance (:doc:`quickstart`). The main difference is that the client needs an authorization token to connect to a federation hosted on FEDn Studio. These are the steps to configure and attach a client to a project: | ||
|
|
||
| In order to generate some data for training, on your local machine, navigate to the mnist-pytorch example: |
There was a problem hiding this comment.
In order to set up the local environment and configure local client data access, ...
docs/studio.rst
Outdated
| bin/split_data | ||
| bin/build.sh | ||
|
|
||
| This commands will generate some training data and two files: package.tgz and seed.npz. The first is a compute package file and the second is a model file that can be used as the inital seed model. (For a more detailed explanation of the commands, see the :doc:`quickstart`) |
There was a problem hiding this comment.
These commands will generate training data..
| @@ -3,6 +3,8 @@ APIClient | |||
|
|
|||
| FEDn comes with an *APIClient* for interacting with the FEDn network. The APIClient is a Python3 library that can be used to interact with the FEDn network programmatically. | |||
|
|
|||
| Connect to the network | |||
| ---------------------- | |||
|
|
|||
| The APIClient is available as a Python package on PyPI, and can be installed using pip: | |||
There was a problem hiding this comment.
Confusing, the APIClient is part of the fedn package, which is installed from pip?
docs/apiclient.rst
Outdated
| @@ -18,4 +20,60 @@ To initialize the APIClient, you need to provide the hostname and port of the FE | |||
| from fedn import APIClient | |||
| client = APIClient("localhost", 8092) | |||
|
|
|||
| Inital setup | |||
There was a problem hiding this comment.
Set the active compute package and the active seed model
docs/apiclient.rst
Outdated
| Inital setup | ||
| ------------ | ||
|
|
||
| Before you can start training models, you need to set the active package and an initail seed model. The active package can be set using the following code snippet: |
There was a problem hiding this comment.
Suggest to delete the sentence "Before you can start..."
docs/apiclient.rst
Outdated
|
|
||
| .. code-block:: python | ||
|
|
||
| client.set_active_package(path="path/to/compute_package.zip", helper="numpyhelper") |
There was a problem hiding this comment.
I think packages need to be tar.gz, tgz ? Plz check.
docs/apiclient.rst
Outdated
|
|
||
| .. code-block:: python | ||
|
|
||
| client.set_active_model(path="path/to/seed_model.zip") |
There was a problem hiding this comment.
They are not usually zipped. Usually ends with ".npz" (numpy compressed)
docs/studio.rst
Outdated
| Studio | ||
| =============== | ||
|
|
||
| FEDn Studio is a web-based tool for managing and monitoring federated learning experiments. It provides a user-friendly interface for monitoring the progress of training and visualizing the results. |
There was a problem hiding this comment.
Maybe add / harmonize Studio intro with features mentioned in the new main README.
docs/studio.rst
Outdated
| Getting started | ||
| --------------- | ||
|
|
||
| Before you can start using Studio, you will need an account. Head over to `studio.scaleoutsystems.com/signup <https://studio.scaleoutsystems.com/signup/>`_ and sign up today! |
There was a problem hiding this comment.
"..sign up today!" -> "sign up."
docs/studio.rst
Outdated
|
|
||
| Once you have created a project, you can find it via the sidebar link Projects. Here you will find the list of all your projects. When inside a project you can see the following tabs in the sidebar: | ||
|
|
||
| 1. **Dashboard**: The dashboard provides an overview of the project. The controller and combiner(s) are listed under "Network". This is also where you can find the current FEDn version and get the option to upgrade to a newer version if available |
There was a problem hiding this comment.
Add punctuation to last sentence in all items.
docs/studio.rst
Outdated
| .. image:: img/studio_project_overview.png | ||
|
|
||
|
|
||
| Configure and attach clients |
There was a problem hiding this comment.
This covers two things, "Initialize the project with a compute package and a seed model", and "connect a client".
docs/studio.rst
Outdated
|
|
||
| Navigate to clients and click on the "Download config" button. This downloads a client config file. Place this file in the mnist-pytorch directory on your local machine. Rename the file to client.yaml. Then edit it to set the name of the client. | ||
|
|
||
| Later, in Studio, click on the "Connect client" button. |
There was a problem hiding this comment.
I think we should update this with "native instructions" also.
There was a problem hiding this comment.
Would like to leave this as is for now... Just until we've settled on a client connect view in studio.
Move Studio docs to Fedn docs
If you have time to test this locally, please do... Use this sphinx command to build: sphinx-build . _build and then run a http server from the _build directory.