Discussions

Dear people, How do you use the image upload in the visual content moderation API? I mean uploading the data, not publicly hosting it and uploading the URL. I tried to generate a request via the API docs: https://docs.thehive.ai/reference/post_v2-task-sync I've downloaded the 1x1 pixel (https://en.wikipedia.org/wiki/File:1x1.png) and uploaded it via the option "media stream". This generates the following requests on the docs page: ---------- curl --request POST \ --url https://api.thehive.ai/api/v2/task/sync \ --header 'Authorization: Token {MY_API_TOKEN}' \ --header 'accept: application/json' \ --header 'content-type: application/json' \ --data ' { "image": "data:image/png;name=1x1.png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABAQMAAAAl21bKAAAAA1BMVEUAAACnej3aAAAAAXRSTlMAQObYZgAAAApJREFUCNdjYAAAAAIAAeIhvDMAAAAASUVORK5CYII=" } ' ---------- However, executing the request results in the following error: ---------- { "return_code": 400, "message": "Expected input to be an object, received: data:image/png;name=1x1.png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABAQMAAAAl21bKAAAAA1BMVEUAAACnej3aAAAAAXRSTlMAQObYZgAAAApJREFUCNdjYAAAAAIAAeIhvDMAAAAASUVORK5CYII= - context: [\"image\"]" } ---------- What is the proper way of using this image upload API? (And could you maybe fix the docs page?) Thanks in advance!

Is there a way that we can request from localhost?

My question related to this section: https://docs.thehive.ai/reference/submit-a-task-asynchronously How can we verify that the request received to our callback endpoint is coming from the Hive?

How do I report an incident? I am facing a problem in production with your API domain (the DNS is routing requests to https://api.thehive.ai/api/v2/task/sync to a down IP, 199.83.232.127:443).

Where i can find monitoring from hive API ?

Hello, I am struggling to create and set up a project in the dashboard. I have created a project. In the first step "settings" under the general tab, I have filled all the required fields. Then I moved to the classes tab. I have added two classes and when I click save, I am receiving "2 label groups approve / reject should be present" message. What should I do?

About text moderation, is there support for Brazilian Portuguese? In supported languages (https://docs.thehive.ai/docs/classification-text#supported-languages), the portuguese language does not have all the matches. Example promotions and spam, and please what means '22 Q1? Sexual Violence Hate Bullying Promotions Spam Pattern Match '22 Q1 Pattern Match '22 Q1 - - Thanks!

Hey there. We're submitting text reviews but are noticing that on short strings (~5 words or less), it sometimes responds with potential profanity (ie. "Joe Dick" returns profanity on the word "Dick"). At first glance, it makes sense, but it is there a recommended minimum amount of words within a string to avoid false positives?

As per your API reference manual, we have to use an API key to authenticate a request. If we call your API from our backend server, there will be no issue. But if we want to use it from a client app (web/mobile applications), there is will be a security issue. is there any way to generate a temporary auth token to use it in web/mobile applications?

What is the cost per text message for the API?