The server API version is greater than the version supported by the activity Issue

Help Document Understanding
1

Hello everyone,

I have issue that I faced when I was working on Document Understanding project.

I have been searching solution for this issue for 2 days and I could not find out any solution.

I deployed a MLSkill on AICenter and I build a project by using DocumentUnderstandingFramework but Classify Document Scope activity is returning an error as;

Retry Scope - Classify: Request CorrelationId: 56774cbe-f686-4920-9343-c226495ed7ba
The server API version is greater than the version supported by the activity.

Here is my package versions;

UiPath.DocumentUnderstanding.ML.Activities v1.13.2
UiPath.IntelligentOCR.Activities v6.0.1

ML Package Version is 22.4.3.1

Here is my public API Url;

https://ai-uipath.deskover.com/public/mlskills/27394d1e-f152-46b9-945b-7b01af1717ea/a91b7fe2-c144-41cf-b67b-0bb1f257ce8d/deployold

MLSkill Logs;

Traceback (most recent call last):
File “”, line 222, in predict
File “”, line 110, in validate_api_version
util.misc.ServiceException: Service Exception 406: InvalidApiVersion - api_version should be one of [‘1.2’]
2022-11-13 16:57:18,415 - uipath_core.plugin:metering_event:319 - INFO: [ correlation_id: c1c87802-56b5-4360-a05c-5251c0d46af3 ] [ prediction_id: ] - Metering-Server-Request-Time: 0.06910157203674316
2022-11-13 16:57:18,416 - uipath_core.plugin:run_model_function:638 - ERROR: Request failed, reason: Traceback (most recent call last):
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/plugin.py”, line 620, in run_model_function
output, metrics = function(*args, **kwargs)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/plugin.py”, line 392, in _model_predict
return self._model_predict_with_context(user_data, **kwargs)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/auth/auth_processing_filter.py”, line 75, in _decorator
return fn(*args, **kwargs)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/logs/logger_mvc.py”, line 91, in _decorator
return fn(*args, **kwargs)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/auth/auth_processing_filter.py”, line 25, in _decorator
return fn(*args, **kwargs)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/plugin.py”, line 436, in _model_predict_with_context
raise result_dict[“predict_exception”]
uipath_core.exceptions.model_exceptions.ModelException: Service Exception 406: InvalidApiVersion - api_version should be one of [‘1.2’]

2022-11-13 16:57:18,417 - uipath_core.controller:handle_invalid_usage:133 - ERROR: {‘info’: ‘Prediction Failed’, ‘reason’: ‘{“message”: “Service Exception 406: InvalidApiVersion - api_version should be one of ['1.2']”, “stacktrace”: " File \“/microservice/main.py\”, line 74, in predict_with_metadata\n response, telemetry = service_internal.predict(context, doctype_name=model_name or self.doctype_name)\n File \“\”, line 222, in predict\n File \“\”, line 110, in validate_api_version\nutil.misc.ServiceException: Service Exception 406: InvalidApiVersion - api_version should be one of ['1.2']"}’}

Can you please help me with that problem?

2

Hi @Ulas_Yilmaz

Why dont you downgrade your DU library to required version?.

Regards,
Anil Bhargav

3

version 1.2 is not available.l
Do you have any suggested versions?

4

Hi @Ulas_Yilmaz

Can you confirm few thinga first… did you run a full pipeline before creating the skill? And what ML are you using is it the classification ML? Or in datalabelling did you indicate the classification label as well? Can you try with any other skill if setup is good?

Which studio version are tou using and what is the compatibility is it wondows or windows- legacy?

Cheers

5

–did you run a full pipeline before creating the skill?
I run evaluation and train pipelines by order. I didn’t run full pipeline.

–what ML are you using is it the classification ML?
It’s Out of the box Packages → UiPath Document Understanding → DocumentUnderstanding

–datalabelling did you indicate the classification label as well?
It’s a custom model so that I couldn’t set datalabelling…

Can you try with any other skill if setup is good?
I tried many skill with ML Package Version; 22.10.1.0, 22.5.2.0, 22.4.3.0 and I get same error with thoose version.

If you have any project developed by document understanding framework, can you please share its version with ML package version that used in this project?

6

Hi @Ulas_Yilmaz

Yes we are using .so here are the details

System - 22.4
Documentunderstandding.ml.activities - 1.13.2

Classifier - 22.4.0.0
Invoice extraction - 22.4.1.0

Even we are using one mode for classification and other for extraction

Cheers

7

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.