-
Notifications
You must be signed in to change notification settings - Fork 68
[SN-92] create a foundry + send to annotate workflow example #1339
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
| @@ -0,0 +1,352 @@ | |||
| { | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could we add some brief text that explains what is the point of this step ?
Something like: This step is meant to generate annotations that can later be reused as pre-labels in a project
and link docs perhaps ? https://docs.labelbox.com/docs/model-runs
Reply via ReviewNB
| @@ -0,0 +1,352 @@ | |||
| { | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we need to explain this a bit better:
Something like: Mapping a model's ontology to a project's ontology is currently not supported through the SDK, however, to showcase how to send foundry predictions to a project, we are going to generate the mapping of the foundry app ontology to the project ontology through the UI.
- Navigate to your dataset you created for your model run
- Select Select all in the top right corner
- Select Manage selection > Send to Annotate
- Specify the project we created from the project dropdown menu
- Selecting a workflow step is not required since we are not sending annotations to a project using this UI modal.
- Mark Include model predictions then scroll down and select Map
- Select the incoming ontology and matching ontology feature for both Computer and Phone
- Once both features are mapped press the Copy ontology mapping as JSON in the top right corner
- Do not save this configuration, since we are not sending predictions to a project using this UI modal. We will be sending predictions in the following steps using the SDK.
Reply via ReviewNB
| @@ -0,0 +1,352 @@ | |||
| { | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should the title use "annotations" or "predictions" ?
- Could we format parameters like this -->
predictions_ontology_mappingplease? predictions_ontology_mappingfor this one, I think we need to clarify, something like, "A dictionary containing the mapping of the model's ontology feature schema ids to the project's ontology feature schema ids.override_existing_annotations_rulefor this one we are missing one resolutionOverrideWithAnnotations = "OVERRIDE_WITH_ANNOTATIONS"Do we know what this is ? and how to use it?
Reply via ReviewNB
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hey! Override with annoations is meant for when you want to override the annotations on a project from one project to another. That is why I did not include since it would not be application for that method. I believe it does the exact same as OverrideWithPredictions when it is with a model. Though it lets me pass in either or and gives the same result. Im sending in some commits now from my changes so far
|
I think we should only include the classification example - I think the object detection and text entities is just redundant |
|
@aovallelb I think maybe the same thing but the object detection one instead. That one shows off the foundry. It picks up 20-30 objects for the cost of being slightly more complicated, which might be a good thing since you have to map your ontology. It also uses an image we use across multiple notebooks, which is a nice touch. But do you think it should be in a new or existing folder? Maybe extras? |
@Gabefire sounds good; I ll review that one. I think it is ok to leave that single file in the foundry folder for now, since, hopefully, there will be more methodologies in the near future that we could create notebooks for, wdyt ? |
|
@Gabefire feel free to incorporate the changes I suggested on the object detection notebook only, and let me know once is ready for review. |
|
@aovallelb Hey I made the changes for the conflictresolution I believe the Overridewithannotations is for if you go from one project to another with the other send_to_annotate method. But when I used it, it would work but do the same thing as overwithpredictions. I have it included in the notebook. Also, just a heads up, there is a bug with the run_foundry_app method. You have to have a description in your foundry app. I actually fixed this on another. PR, and they approved it, but when you go to test, you will likely need to include a description |
| "source": [ | ||
| "## Step 6: Map ontology through the UI\n", | ||
| "\n", | ||
| "Something like: Mapping a model's ontology to a project's ontology is currently not supported through the SDK, however, to showcase how to send foundry predictions to a project, we are going to generate the mapping of the foundry app ontology to the project ontology through the UI.\n", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you remove "Something like:" - Once removed this all looks good and you can merge!
Finished notebook on foundry SDK basic methods. I made a new folder on this since no folder seemed to match foundry. Also, not sure if you want me to do other examples besides just images i.e. make a notebook on text models