DALL-E (Model Plugin)
Last updated
Last updated
The DALL-E (Model Plugin) plugin allows you to generate images. The generated images will appear as attachment to the asset where they were run.
As part of the prompt sent to DALL-E (Model Plugin), you may choose to send the entire text of the asset if the asset is text, the text that was highlighted as part of an entity annotation, or custom text written by a user into a Text classification box.
From the Plugin Directory, search for DALL-E (Model Plugin) and install the plugin to your organization. More information on installing plugins can be found in the page.
The name of the plugin is DALL-E (Model Plugin) and the creator of the plugin is onur (at) imerit.net.
Navigate to the project where you'd like to use ChatGPT.
If you'd like for DALL-E (Model Plugin) to automatically receive and process tasks, you may integrate it into your workflow.
From the Workflow tab in your project, drag a Plugin stage into the workflow view, and plug it where you'd like it to be. Then, click on the stage you've just placed and select "DALL-E (Model Plugin)" from the list of models available.
In the example above, DALL-E (Model Plugin) will receive tasks from the Start stage, process them, and send them out to a labeling stage. More info on the other settings in the plugin settings panel can be found below.
openai_api_key: By default, the DALL-E plugin will use an OpenAI API key provided by iMerit, for testing purposes, limited to generating, globally, 5 images per minute. You may provide here your own OpenAI API key to remove these limits and utilize your own OpenAI API key.
model_name: When providing a custom OpenAI API key, you may indicate here the name of the model the API key is for. Currently, OpenAI provides dall-e-2 and dall-e-3.
Please ensure the OpenAI API key you provide belongs to the model you specify.
For example, providing an OpenAI API key for dall-e-2, and then selecting dall-e-3 as the model will cause the plugin not to work.
get_prompt_from: Where DALL-E should look for when it searches for the prompt.
Options: text-annotation, entity-annotation, or full-text-asset. More information on prompting below.
image_size: Resolution of the image to generate. Recommended: 512x512.
You may run the DALL-E (Model Plugin) plugin to generate images directly from the labeling editor. After having added DALL-E (Model Plugin) to your organization, enter a project, and click on the "Plugins" icon on the top toolbar. The DALL-E (Model Plugin) plugin will be available in the dropdown.
In the Config JSON field, ensure the get_prompt_from property is set to full-text-asset.
The text of the asset will be passed as the prompt.
You may change the image_size property of the Config JSON to pick an image resolution.
To pass the entire text of the asset as a prompt, first ensure that the asset is a .txt file.
Map DALL-E (Model Plugin)'s Prompt class to the class you'd like to pass as prompt to DALL-E (Model Plugin). Text highlighted in text assets with this class will be the prompt.
In the Config JSON field, ensure the get_prompt_from property is set to entity-annotation.
Map DALL-E (Model Plugin)'s Prompt class to the class you'd like to pass as prompt to DALL-E (Model Plugin). Text typed in this text classification tool will be part of the prompt.
In the Config JSON field, ensure the get_prompt_from property is set to text-annotation.
If you have plugged DALL-E (Model Plugin) into your workflow, DALL-E (Model Plugin) will be activated every time a task is passed into its input plug.
Click on the DALL-E (Model Plugin) text to run the plugin with its default preset (if any, ). Otherwise, click on the three dots to the right to run the plugin with custom settings. The plugin settings dialog will appear. More info on the plugin settings that can be found in this dialog are below.
If you wish to run DALL-E (Model Plugin) directly from the labeling editor, follow the instructions .
You may check the progress of the conversion from the Plugin Sessions dialog. More information on checking plugin progress .
