ai-102-ai-engineer-assoc video for calling a Custom Vision model prediction endpoint over HTTP requires three pieces of information. [A]: This key has to be set
Calling a Custom Vision model prediction endpoint over HTTP requires three pieces of information. [A]: This key has to be set as a header in all requests. That's what gives us access to the endpoint. [B]: The dialog shows two different URLs. If we're posting an image URL, then use the first URL, which ends in /url. If we want to post a raw image in the body of our request, we use the second URL, which ends in /image. [C]: If we're posting a raw image, we set the body of the request to the binary representation of the image and the content type to application/octet-stream. If we're posting an image URL, we put that as JSON in the body and set the content type to application/json.