简体   繁体   中英

AWS Sagemaker Unable to Parse Augmented Manifest File

I've read all of the existing documentation on Augmented Manifest files. I see no difference from my file, but I keep experiencing this error when training:

ClientError: Data download failed:Failed to download data. Unable to parse augmented manifest, error in line: 1

My example first line:

{"source-ref": "s3://test-bucket/test-data/test/bucket/10done.png", "video-frame-object-detection": {"annotations": [{"class_id": 1, "top": 880, "left": 43, "width": 2499, "height": 324}], "image_size": [{"width": 2543, "height": 2543, "depth": 3}]}, "video-frame-object-detection-metadata": {"class-map": {"0": "Good", "1": "Bad"}, "human-annotated": "no", "creation-date": "2022-06-09T11:01:27.440682", "type": "programmatically-created-labels"}}

These are how my breaks look (end of file):

![在此处输入图片描述

How I manually created the manifest file: ![在此处输入图片描述

Sagemaker is recognizing my attributes as well.

The error is clearly coming from the syntax error from your Augment manifest files. It should be in jsonlines format. Can you check with some gener

The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM