Hi Team,
I am trying to create a project on Invoice Extraction using DU.
I have already done the data set label and Tried to create an extractor as well as pipeline training run via aicenter but it always fails with the same error. There is no update on the forum or documentation about the error.
This is going to be a duplicate one, but somehow I can’t update the last topic as it is resolved already.
The pipeline training run failed with the below error.
AI_Center_Issue: Failed to upload file:
2023-10-06 10:49:30,335 - uipath_core.training_plugin:save_model:164 - INFO: Model save successful with response None
2023-10-06 10:49:30,335 - uipath_core.training_plugin:run_train_only:273 - INFO: Total time taken for trainer execution, time: 325.95547556877136
2023-10-06 10:52:01,245 - uipath_core.storage.azure_storage_client:upload:140 - ERROR: AI_Center_Issue: Failed to upload file: PipelineRun/Pipeline_35335_4b92628c657345bbaeabbde054b718ef/1/KodakExtractor.zip to bucket: training-1d34503e-6b6c-4672-9356-16f049d7da5f-train-artifactsv2, error: The specified container does not exist.
RequestId:8aabd4c4-401e-0030-5943-f84018000000
Time:2023-10-06T10:52:01.2642031Z
ErrorCode:ContainerNotFound
Error:None
Content: <?xml version="1.0" encoding="utf-8"?><Error><Code>ContainerNotFound</Code><Message>The specified container does not exist.
RequestId:8aabd4c4-401e-0030-5943-f84018000000
Time:2023-10-06T10:52:01.2642031Z</Message></Error>
2023-10-06 10:52:06,250 - uipath_core.utils.utils:_retries:194 - WARNING: Function: upload execution failed, retry count 1
2023-10-06 10:52:06,270 - uipath_core.storage.azure_storage_client:upload:140 - ERROR: AI_Center_Issue: Failed to upload file: PipelineRun/Pipeline_35335_4b92628c657345bbaeabbde054b718ef/1/KodakExtractor.zip to bucket: training-1d34503e-6b6c-4672-9356-16f049d7da5f-train-artifactsv2, error: The specified container does not exist.
RequestId:8aabe699-401e-0030-4e43-f84018000000
Time:2023-10-06T10:52:06.2931685Z
ErrorCode:ContainerNotFound
Error:None
I tried multiple runs, My file size is minimal… I am following the naming convention yet this is not working.
I am using OOB-> UiPath Document Understanding-> Invoices
The dataset is correct INFO: Model save successfully
It’s nothing with the dataset but rather an issue with the container.
The dataset is as per standard documentation. I have created 10-15 pipelines before in enterprise license and it never failed. Currently, I am on the community version of a POC.
Please look into detailed logs may be for one of the provided files its not trained or some issue with input may be…as once it is returning null for training
Hi @lakshay.verma
The container error indicates that the container of the ML could not be found.
The container encapsulates the traning environment and code and libraries…
I have some questions to understand more the problem.
-Did u upload ur dataset and train extractor via Document understanding or IA CENTER ?
-the pipeline is genrated by DU or u created a manual pipeline?
Can u share more details about the steps you used.
Did u upload ur dataset and train extractor via Document understanding or IA CENTER ?
I tried it both ways, first with Document Understanding and later with AI, however, the results were the same.
the pipeline is generated by DU or u created a manual pipeline
Also for this, I tried both ways but the results were similar. I tried creating a pipeline from the AI by creating pipeline and using the dataset exported from DU and creating a training pipeline which failed with the same error later within DU I tried to create the extractor both ways with manual training and automated training but the results were the same.
I was under the impression that the issue was with dataset size but that’s not a scenario as I downsized my sample data.
The dataset generated by DU has a specific structure. In addition to images or PDFs, it contains JSON files with a document schema. So, if you train the pipeline on the dataset generated by DU, it can lead to errors.
Container errors may be related to the name and format of the dataset or the files.
If you upload a dataset via DU, you will receive confirmation that your dataset is okay, and you will be alerted if a specific file doesn’t match the criteria.
Therefore, I recommend that you delete the dataset, package, and ML skill, and then repeat the entire process, either via DU or IA Center, to avoid any confusion between generated and manually added elements.
I see the problem was with my account and not the steps.
I replicated the same steps on my other personal account and followed the same process and it went through. Again, I am not sure what’s the issue with my current account but it definitely not working.
I could not find any solution for the above. I tried all permutations and combinations but things are not working as expected. I picked up another account and ran it through the same steps and it went through.
ashokkarale
jast1631
Yoichi
lrtetala
singh_sumit
Anil_G
pradeep931
once10201029
parnalmahavir.patni
Botzilla