Add tutorials on project and artifacts (Closes #1680) #1683
Conversation
brollb
left a comment
brollb
left a comment
There was a problem hiding this comment.
I made a few edits. This reads a bit more like an overview of the UI whereas I was expecting it to describe some of the basics while working on a simple example project (like in the demo yesterday). For example, rather than having a section for "Managing Projects", it might be better to have something like "Hello DeepForge!" where it creates a project and builds a "hello world" like pipeline to run. Presenting other areas of the UI where projects can be created can be removed.
For the "Managing Data in DeepForge" section, I would prefer reframing this to be working on an example as well (perhaps extending the "Hello DeepForge!" example). That said, creating the pipeline to visualize the mnist images is better as it is a bit more project-focused. However, it would be better to have this step as just one step in the tutorial. For example, you could start with the mnist example, train a network and then show misclassified images alongside their predicted labels.
There are also a few grammatical errors that are addressed in the following inline comments.
docs/tutorials/storage.rst
Outdated
| Loading a pickled file | ||
| ---------------------- | ||
|
|
||
| Let's load in a pickled file and visualize the contents of the file in `DeepForge`. Use the following ``code`` to generate the pickle file ``mnist_sample.pkl``. |
There was a problem hiding this comment.
Why isn't this an operation in deepforge?
| def execute(self, dataset): | ||
| images = dataset['X'] | ||
| for i in range(9): | ||
| # define subplot |
There was a problem hiding this comment.
What value does this comment add?
| Visualizing Images | ||
|
|
||
|
|
||
| Visualizing a FITS file with custom serialization |
There was a problem hiding this comment.
This should be expanded upon and then moved to its own example/tutorial.
| .. figure:: custom_utils.png | ||
| :align: center | ||
|
|
||
| Custom Utils |
There was a problem hiding this comment.
Something like "Custom Python Modules" or "Custom Python Utilities" would be better...
|
|
||
| Custom Utils | ||
|
|
||
| After you click on Enter, a text editor will open. Copy and paste the following content: |
There was a problem hiding this comment.
I generally would not recommend saying things like "Copy and paste the following content" as the user can probably figure it out... It's generally better to stay focused on the content of the project rather then the ways that people follow along.
|
|
||
| The `pipeline` has a single operation called `VisualizeFITS` that uses the FITS file we uploaded with custom serializer. | ||
|
|
||
| Using data available different storage backends |
There was a problem hiding this comment.
This doesn't really make sense. I suspect you are referring to importing existing data from a storage backend...?
|
|
||
| Importing Existing Data | ||
|
|
||
| After you click on Import Existing Data, a Modal as shown below will pop up, select `Sciserver Files Service` from the dropdown menu. Provide your `username` and `password` for SciServer and select correct volume pool and file path as shown in the figure below: |
There was a problem hiding this comment.
- "Import Existing Data" should be either quoted or italicized since you are referring to it as a noun and not a verb phrase.
- "Modal" should not be capitalized.
- "select the correct volume pool"
|
|
||
| SciServer File Import | ||
|
|
||
| Upon Success, the artifact will be available for use in DeepForge. |
There was a problem hiding this comment.
Why is "Success" capitalized?
|
As this is now being added by @KyleAMoore, in #1740. I am closing this. |
No description provided.