In Design for aiXplain, you can create pipelines that combine different AI functions to transform your data and build your own AI application. Nodes are the components that represent these functions in your pipeline. You can connect nodes to each other to define the order and the logic of your data processing. Depending on the type of input and output data, and the type of function, there are five types of nodes that you can use in Design:
Input node: This node allows you to provide the input data for your pipeline. The input node will feed the input data to the next node in your pipeline.
Output node: This node allows you to receive the output data from your pipeline. The output node will receive the output data from the previous node in your pipeline.
Decision: This node allows you to create if/else statements in your pipeline. You can set a condition based on the input data or the output of another node, and connect the true/false branches to different nodes in your pipeline. This node is useful for creating conditional logic or branching paths in your pipeline.
Router: This node allows you to route your data to different nodes based on their data type. You can define the data types, such as text, audio, image, or video, and connect them to the corresponding nodes in your pipeline. This node is useful for handling heterogeneous or multimodal data in your pipeline.
Model node: This node allows you to add different types of functions to your pipeline. You can choose from four function types: All, AI, Segmenter, Reconstructor, or Utility. Depending on the function type, you will see different options for the function dropdown. The function dropdown lists the model functions from the Discover marketplace that perform various AI tasks, or the operations that segment, reconstruct, or extract data.
Segmenter/Reconstructor: These nodes are used to specify the size of the segments passing through the pipeline. The segmentor node splits the input data into smaller segments, while the reconstructor node merges the output segments into a larger one.
Utility Function: If you select the utility function type, you will see the option "extract audio from video" in the function dropdown. This option will take a video file as input and produce an audio file as output. You can use this option to isolate the audio track from a video file.
Metric node: This node allows you to measure and compare the performance of different models or pipelines based on various metrics, such as quality, latency, footprint, cost, and bias. You can select the metrics that are relevant for your AI function and view the results.
I hope this article has proven to be both helpful and informative for you. We greatly appreciate your decision to select aiXplain as your AI creation and optimization partner. Should you have any inquiries or feedback, please don't hesitate to reach out to us at your convenience.
Contact us at: firstname.lastname@example.org
Was this article helpful?
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
We appreciate your effort and will try to fix the article