I've followed these tutorials:
Amazon's - https://docs.aws.amazon.com/sagemaker/latest/dg/sms-getting-started.html
Medium.com - https://medium.com/#bruceyanghy/end-to-end-guide-for-setting-up-aws-sagemaker-ground-truth-public-data-labeling-jobs-8732cf49b865
I'm attempting to run a basic object detection (images with bounding boxes) Ground Truth labeling job with Mechanical Turk. I've tried multiple times, opening up any and all security on the S3 bucket containing the images. My past two attempts have been the closest, but still only manage 32/389 images labeled (357 Failed) and 62/389 images labeled (327 Failed). The images are exactly the same in both attempts.
Checking the manifest file for clues, the only error I get is:
<Code>InvalidArgument</Code>
<Message>Requests specifying Server Side Encryption with AWS KMS managed keys require AWS Signature Version 4.</Message>
<ArgumentName>Authorization</ArgumentName>
<ArgumentValue>null</ArgumentValue>
I can't find any references anywhere as to what I need to change to make this work. I've tried with and without server-side encryption on the S3 bucket.
How do I change this signature version?
Related
I'm having some trouble recovering from failures in attaching custom images to my sagemaker domain.
I first created a custom image according to here.
When I use sagemaker console to attach the image built with sm-docker, it appears to successfully "attach" in the domain's image list, but when inspecting the image in the console, it shows an error:
Value '' at 'appImageConfigName' failed to satisfy constraint: Member
must satisfy regular expression pattern
This occurs even when the repository or tag are comprised of only alphanumeric characters.
After obtaining this error, I deleted the repositories in ECR.
Since then, the domain fails to update and I am unable to launch any apps or attempt to attach additional images.
The first issue I would like to address is restoring functionality of my sagemaker domain so I can further troubleshoot the issue. I am unable to delete the domain because of this error, even when there are no users, apps, or custom images associated with the domain.
The second issue I would like to address is being able troubleshoot the appImageConfigName error.
Thanks!
While I was unable to delete the domain via console, I was able to delete it via cli.
I'm trying to do a quick PoC on the AWS Rekognition custom labels feature. I'd like to try using it for object detection.
I've had a couple of attempts at setting it up using only tools in the AWS Console. I'm using images imported from the Rekognition bucket in S3, then I added bounding boxes using the tools in the Rekognition console.
All of my images are marked up with bounding boxes, no whole image labels have been used. I have 9 labels, all of which appear in at least 1 drawing.
I've ensured my images are less than 4096x4096 in size (which is mentioned on this AWS forums thread as a possible cause of this issue.
When I attempt to train my model I get the "The manifest file contains too many invalid data objects" error.
What could be wrong here? An error message complaining about the format of a file I didn't create manually, that I can't see or edit isn't exactly intuitive.
Unknowingly I have deleted the below buckets from my project
artifacts.<PROJECT-ID>.appspot.com
us.artifacts.<PROJECT-ID>.appspot.com
This has deleted all the images from gcr. Let me know if the above buckets are where the gcr images are stored or is it something else?
Also when I created a new image and pushed it to gcr, all the deleted images in gcr console got recovered. But whenever I try to pull any old image it is throwing "unknown blob" error.
Yes, these buckets are where the docker container artifacts are built and stored..(Artifacts being the build steps results, that add up to an image)
Then they are referenced by the Google Container Registry (i.e. gcr.io), but they will be still located in your bucket.
Since you removed the bucket and its contents, you will be missing old building steps from your built images, that's why you get the error pulling image configuration: unknown blob error message.
For example, I uploaded a new image following this documentation, and I removed the artifacts.<PROJECT-ID>.appspot.com bucket afterwards. Then I reuploaded it, using a tag (I used quickstart-image:tag1, and when pulling it this way:
docker pull gcr.io/wave16-joan/quickstart-image:latest
I got the error pulling image configuration: unknown blob error message, because it's missing the steps I already had in my previous build.
However, doing this:
docker pull gcr.io/wave16-joan/quickstart-image:tag1
Allowed me to pull the image, but the image wasn't usable.
Regarding your second question, I believe that the reason why you are seeing in the Container Registry references to the images you removed, it's because GCR is still saving the references to the steps from building these images, however since they are deleted, they are not able to be pulled.
I've deployed the AWS Cloud formation Serverless image handler to resize images in an S3 bucket which is working well. It is using Cloudfront, API Gateway, Lamdba, and a custom domain.
Hosted within the same bucket are audio files. Now when I hit the Cloudfront URI for the audio files I receive at 502 error with a JSON response "Internal server error". Original image paths continue to work well.
I'd like to ignore (one way or another) all non-image files. One way that I have been attempting is to add a prefix to the GET request for media files. eg: /media/{s3-key}. But I am having difficulty getting that working and am thinking there must be another way.
For example: why would there be a 502 error if the image processing prefixes (/fit-in/400x400/{s3-key} for example) not be present? Why would Lambda be trying to process those rather than just returning?
Or even if the request makes it to Lambda: why fail rather than returning the file?
I'm finding the right way to use AWS Rekognition service.
My problem is How to verify a person image on multi collections, I'm reading Build Your Own Face Recognition Service Using Amazon Rekognition | AWS Machine Learning Blog from Amazon but cannot find the implementation document for it. My point is Face verification title.
Update 1:
My target is: Using AWS Rekognition to get person's info by their face.
My problem is: How to make AWS Rekognition improves its accuracy when recognizing a face.
What I tried:
Upload multi captured portraits of a person with same ExternalImageID but I'm not sure it works or not.
Finding a way to create Collection for each person, then upload person's portraits to their Collection but I don't how to search a face through multiple Collections.
I'm trying use S3 for storage people's images then using Lambda function to do something that I've not got yet.
Update 2:
What is your input material: Input materials are some people's portrait photo with ExternalImageID is their name (eg: my portrait photo will have ExternalImageID is "Long").
What are you trying to do: I'm trying to get ExternalImageID when I send a portrait photo of a registered person. (eg: with my other portrait photo, AWS has to response ExternalImageID is "Long").
Do you have it working, but it is not recognizing some people? Yes, it's work but sometimes it cannot recognize exactly people.
Please tell us your use-case / scenario and what you are trying to accomplish:
Create an AWS Rekognition collection with sample name (eg facetest).
Register some people with their name is ExternalImageID.
Submit an image to AWS Rekognition API to get ExternalImageID - his name.
Okay, so basically you have it working but it doesn't always recognise the person. I'll assume it does not even list the person in the response, even with a low percentage.
I would recommend adding multiple images of the same person to the Face Collection, specifying the same ExternalImageId for each image. (Use one Face Collection with all people in it, including multiple images of the same person.)
Please note that "If you provide the same image, specify the same collection, and use the same external ID in the IndexFaces operation, Amazon Rekognition doesn't save duplicate face metadata." However, adding different images with the same ExternalImageId should be fine.