AI Center | Data Extraction Scope Error | Custom ML Model

Help AI Center
,
1

Hi Team,

I have created Custom ML Model and using the same trained model as in the configure extractors have added the fields which I created in the data labeling model. I am getting an error. Can you tell me what the issue here is?

Data Extraction Scope:

image
image542×619 15.6 KB

Configure Extractors:

image
image901×585 143 KB

Document Manager:
image

Error:
Data Extraction Scope: Invalid response: content=Service Exception 400: InvalidDoctype - Unknown doctype du. Expected one of [‘invoices’, ‘receipts’, ‘purchase_orders’, ‘utility_bills’, ‘invoices_india’, ‘invoices_au’, ‘passports’, ‘id_cards’, ‘w9’, ‘w2’, ‘delivery_notes’, ‘remittance_advices’, ‘invoices_japan’, ‘invoices_china’]. code=BadRequest trace_id=

2

Hi @SrenivasanKanna

May i know what is the document type that is used as the Input.

Thanks.

3

it’s the certificate of license and document type (Pdf)

4

Hi @SrenivasanKanna

What was the ML Model used to train ?

Thanks.

5

Document Understanding

image
image757×45 964 Bytes

image
image1031×319 13.2 KB

6

Hi @SrenivasanKanna

In Configure Extractors , Click on Icon
image

image

And then Click on “Get Capabilities” → then reassign all the Fields.

Re-run the process and let me know.

Thanks.

7

@SrenivasanKanna Can you check if the model able to classify the document. Try checking with present classification station

8

Thank you so much for your response. I tried and was not able to get the fields it’s throwing below error.

An error occurred while retrieving capabilities from the server. Please check the logs for more details.

9

Yes , I can able to classify the document , only facing issues at the place of extraction.

10

@SrenivasanKanna Did you check the ML skill is up and running. If you are giving the right ML skill though it is not working that means the model not able to communicate. This might resolve some times automatically . If you are using licensed one, then you might have to reach out to UiPath they will install the updated AI Center in your machine

11

Hi @SrenivasanKanna

Please try re-pasting the API key once again and try .

12

I tried, it’s not working , Do I need to run the pipeline ?

13

Hi @SrenivasanKanna

Is the Pipeline run was success if so there is no required to re-run the pipeline.

14

I ran it first time now, waiting for the successful run status.

15

Hi @SrenivasanKanna

Once the Status is Success , Please Update the ML Skill to the Latest Version and the status of the ML Skill should be Available.

Once the above steps are done, please try “Get Capabilities” and map all the Fields Required.

Run the Process.

Thanks.

1 Like
16

Sure, Suraj . I will do the same and update you the status. Thank you so much for your help.

17

Hi Suraj, once again thank you for your help. I ran the pipeline but it got failed. I have added the log below. Any guess?

Error Details: Pipeline failed due to ML Package Issue

2022-04-20 05:43:51,281 - uipath_core.trainer_run:main:73 - INFO: Starting training job…
2022-04-20 05:43:57,759 - uipath_core.storage.azure_storage_client:download:105 - INFO: Dataset from bucket folder training-f452050f-c1ab-412c-b32e-6096da0581e8/df35286d-9c2c-4c7e-b4ae-4fb49ab9d4ef/d4f56c06-c6ea-48ac-8647-3d74cadb4a34 with size 49 downloaded successfully
2022-04-20 05:43:57,760 - uipath_core.training_plugin:train_model:114 - INFO: Start model training…
2022-04-20 05:43:57,760 - uipath_core.training_plugin:initialize_model:108 - INFO: Start model initialization…
2022-04-20 05:43:57,762 - root:_valid_doctype_folder_structure:89 - ERROR: schema.json is empty / does not exist for du dataset
2022-04-20 05:43:57,762 - uipath_core.training_plugin:model_run:150 - ERROR: Training failed for pipeline type: TRAIN_ONLY, error: Document type du not valid, check that document type data is in dataset folder and follows folder structure.
2022-04-20 05:43:57,772 - uipath_core.trainer_run:main:90 - ERROR: Training Job failed, error: Document type du not valid, check that document type data is in dataset folder and follows folder structure.
Traceback (most recent call last):
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/trainer_run.py”, line 85, in main
wrapper.run()
File “/microservice/training_wrapper.py”, line 57, in run
return self.training_plugin.model_run()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 151, in model_run
raise e
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 143, in model_run
self.run_train_only()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 212, in run_train_only
self.train_model(self.local_dataset_directory)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 116, in train_model
self.model.train(directory)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 104, in model
self.initialize_model()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 110, in initialize_model
self._model = train.Main()
File “/microservice/train.py”, line 40, in init
self.opt = self.get_options()
File “/microservice/train.py”, line 127, in get_options
opt = preprocess.configure_options(opt)
File “”, line 130, in configure_options
Exception: Document type du not valid, check that document type data is in dataset folder and follows folder structure.
2022-04-20 05:44:56,130 - uipath_core.trainer_run:main:73 - INFO: Starting training job…
2022-04-20 05:44:59,644 - uipath_core.logs.upload_log_service:upload_logs_file:87 - INFO: Retry Training Triggered:
2022-04-20 05:45:01,376 - uipath_core.storage.azure_storage_client:download:105 - INFO: Dataset from bucket folder training-f452050f-c1ab-412c-b32e-6096da0581e8/df35286d-9c2c-4c7e-b4ae-4fb49ab9d4ef/d4f56c06-c6ea-48ac-8647-3d74cadb4a34 with size 49 downloaded successfully
2022-04-20 05:45:01,377 - uipath_core.training_plugin:train_model:114 - INFO: Start model training…
2022-04-20 05:45:01,377 - uipath_core.training_plugin:initialize_model:108 - INFO: Start model initialization…
2022-04-20 05:45:01,378 - root:_valid_doctype_folder_structure:89 - ERROR: schema.json is empty / does not exist for du dataset
2022-04-20 05:45:01,378 - uipath_core.training_plugin:model_run:150 - ERROR: Training failed for pipeline type: TRAIN_ONLY, error: Document type du not valid, check that document type data is in dataset folder and follows folder structure.
2022-04-20 05:45:01,386 - uipath_core.trainer_run:main:90 - ERROR: Training Job failed, error: Document type du not valid, check that document type data is in dataset folder and follows folder structure.
Traceback (most recent call last):
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/trainer_run.py”, line 85, in main
wrapper.run()
File “/microservice/training_wrapper.py”, line 57, in run
return self.training_plugin.model_run()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 151, in model_run
raise e
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 143, in model_run
self.run_train_only()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 212, in run_train_only
self.train_model(self.local_dataset_directory)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 116, in train_model
self.model.train(directory)
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 104, in model
self.initialize_model()
File “/home/aicenter/.local/lib/python3.9/site-packages/uipath_core/training_plugin.py”, line 110, in initialize_model
self._model = train.Main()
File “/microservice/train.py”, line 40, in init
self.opt = self.get_options()
File “/microservice/train.py”, line 127, in get_options
opt = preprocess.configure_options(opt)
File “”, line 130, in configure_options
Exception: Document type du not valid, check that document type data is in dataset folder and follows folder structure.

18

Hi @SrenivasanKanna ,

Could you let us know what Folder Path have you Chosen as the Dataset for Running the Pipeline ?

19

Hi @SrenivasanKanna

Please check the Folder you have passed as the Dataset for the Pipeline run contains “Schema.jason” file in it.

Thanks.

20

where do I need to the see folder and Schema.jason
I am using Cloud enterprise, any guess on this screen.

image
image1497×717 57.1 KB

@supermanPunch