I am using IBM Watson Visual Recognition for a custom model. I have uploaded my dataset as .zip files, which is fine so far. However, I cannot test already trained model and I cannot label new images. When I go on my Watson services, it says:
Error fetching custom collections: Error in Watson Visual Recognition service: Recieved invalid status 403 in getAllCollections response for guid crn:v1:bluemix:public:watson-vision-combined:us-south:a/649b0335a5a44f6d80d1fd6909e466f9:8a71daa3-b0be-42ac-bb72-1473de835c19:: at endpoint https://gateway.watsonplatform.net/visual-recognition/api/
When I try to label new images , it says: Error in Watson Visual Recognition service: Received empty image data response
To the best of my knowledge, I have checked Google and StackOverflow for solutions, but didn't find any. I am using the Lite version. I only have one project, and one Visual Recognition instance. Please note that it worked for a different Visual Recognition model before, but later I could not use or access that model.
403 is a forbidden error, implying that you don't have authority to access the custom collections. Which is strange as its part of the V4 API available to lite accounts.
If you are doing everything inside Watson Studio, then Watson Studio is accessing the service on your behalf. Watson Studio may have expired its access token. Try flushing your browser cache, logging back into Watson Studio and trying again.
If that doesn't work, you should raise a ticket against Watson Studio.
You can verify if it is a Watson Studio problem by using cURL to list the customised collections - https://cloud.ibm.com/apidocs/visual-recognition/visual-recognition-v4#list-collections
curl -u "apikey:{apikey}" "{url}/v4/collections?version=2019-02-11"
If you still get a 403, then the problem lies with the service. If the cURL command works, then the problem lies with Watson Studio, and how you are using it.
Another thing to consider, since you mentioned you were using a lite plan, is that once you have exhausted the monthly quota of resources for this free plan, you will get 403 error codes.
Related
I am following the tutorial here (https://cloud.google.com/ai-platform/training/docs/training-jobs#python_1) and using Python to deploy a machine learning model to Google cloud for training. However, I got HttpError 403: "Permission denied on resource project my_project." The reason for the error is "CONSUMER_INVALID." Specifically, the returned metadata shows that the consumer is "projects/my_project," and the service is "ml.googleapis.com".
I went to the AI platform and my project is indeed there. The API I am requesting to use is enabled for the project and the environment variable GOOGLE_APPLICATION_CREDENTIALS has been set correctly. I am wondering how to fix the problem. Thanks!
I ran into this issue when I used my literal Project Name. I had to instead use the 'ProjectID' which is listed in the 'Home' Dashboard, in the 'Project info' widget.
I have tried to log out and log in. Deleted the service and created a new one. No success thou. When I look at the url I see I'm being thrown around the entire planet...
Try and log into Watson Studio directly at: https://dataplatform.cloud.ibm.com/settings/regions?context=cpdaas and then select the region where you have provision the Watson Studio instance in IBM Cloud.
Please note that the Watson Studio instance in IBM Cloud is used for entitlement checking and billing. Deleting this instance via IBM Cloud does not remove your Watson Studio projects etc.
I'm trying to push an API that I created in SoapUI into Bluemix using the cloud foundary console. I first created a war file as per this blog: http://alleyfield.com/?p=51
I've managed to get the API into Bluemix, but the results aren't pleasing. In SoapUI I've tested my API and can confirm that it works perfectly. Surely there is a better way of creating an API in Bluemix having created it in SoapUI (soap/xml format.) My problem is that I don't see any options on the Bluemix page that allows for this. I've consulted the official documents, but they almost exclusively focus on sample application APIs using swagger or similar.
I've contacted IBM and they suggested I post this to Stack Overflow as the blog I mentioned is old and doesn't match Bluemix at present.
I completed the basic setup for Google Analytics and inserted the tracking code into my webpage(s). I can log in to analytics.google.com and see historical data just fine.
I have an admin page on my website where I want to display Google Analytics information to several users automatically (no additional logon req'd) and read that this is possible using the Embed API with Server-side Authorization.
I followed the instructions and created a Service Account that shows up under Service account keys in my Google API Manager portal. I also successfully ran 'sudo pip install --upgrade google-api-python-client' on my Cloud9 IDE and pushed the change to my hosting provider, Firebase, i.e. 'firebase deploy'. Lastly, I located and pasted my Analytics View ID in to the code sample provided with the Demo/Setup. My page is served with <!DOCTYPE html> declared.
I can't get it to work. Browser console says:
Uncaught TypeError: window.google.load is not a function
I am thinking that I am not invoking the get_access_token() in that Python module. In other words, I am thinking that server-side scripting is not supported with Firebase hosting.
Anyone know if and how to get Google Analytics Embed API with Server Side Authorization running with Firebase hosting?
This is not exactly a direct answer to your question, but rather an alternative solution to this problem. At least I’m using approach.
You can use Google Design Studio to build and embed amazing analytics and BI dashboards anywhere you like (also on your Admin website). But even easier would be to use Google’s authentication (via gmail) to provide access to select dashboards directly, as Google’s authentication and authorisation is likely more secure that your website’s.
Node-Red on bluemix provides a Watson Machine Translation node. Bluemix recently changed the translation APIs that this uses, releasing a new Watson Language Translation API. (see https://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/doc/language-translation/migrating.shtml for details)
I think this is the reason I used to get an error message in Node-Red saying that the Machine translation could not be binded.
Could you please help?
We're currently working on updates to support the API changes in the nodes. Hopefully, we'll have this working by the end of the week.
This issue has now been resolved. Please update your source and try again...